Discharge_Prediction_Timeseries.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "f4aa2a2b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 230ms/step - loss: 0.8887 - mean_absolute_error: 0.7776 - val_loss: 0.7887 - val_mean_absolute_error: 0.6860\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 194ms/step - loss: 0.5029 - mean_absolute_error: 0.5529 - val_loss: 0.5611 - val_mean_absolute_error: 0.5170\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.3165 - mean_absolute_error: 0.4408 - val_loss: 0.4277 - val_mean_absolute_error: 0.4563\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.2538 - mean_absolute_error: 0.3796 - val_loss: 0.4809 - val_mean_absolute_error: 0.4639\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.1986 - mean_absolute_error: 0.3231 - val_loss: 0.3820 - val_mean_absolute_error: 0.4363\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.1726 - mean_absolute_error: 0.2924 - val_loss: 0.3534 - val_mean_absolute_error: 0.3879\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 194ms/step - loss: 0.1276 - mean_absolute_error: 0.2560 - val_loss: 0.2952 - val_mean_absolute_error: 0.3405\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0943 - mean_absolute_error: 0.2126 - val_loss: 0.2490 - val_mean_absolute_error: 0.3149\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0738 - mean_absolute_error: 0.1902 - val_loss: 0.2182 - val_mean_absolute_error: 0.2872\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 197ms/step - loss: 0.0592 - mean_absolute_error: 0.1673 - val_loss: 0.1919 - val_mean_absolute_error: 0.2638\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0530 - mean_absolute_error: 0.1528 - val_loss: 0.1695 - val_mean_absolute_error: 0.2373\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0385 - mean_absolute_error: 0.1317 - val_loss: 0.1686 - val_mean_absolute_error: 0.2273\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0302 - mean_absolute_error: 0.1143 - val_loss: 0.1308 - val_mean_absolute_error: 0.2024\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0266 - mean_absolute_error: 0.1058 - val_loss: 0.1291 - val_mean_absolute_error: 0.1894\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0237 - mean_absolute_error: 0.0991 - val_loss: 0.1337 - val_mean_absolute_error: 0.2099\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0207 - mean_absolute_error: 0.0954 - val_loss: 0.1472 - val_mean_absolute_error: 0.1902\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0186 - mean_absolute_error: 0.0876 - val_loss: 0.1314 - val_mean_absolute_error: 0.1811\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0153 - mean_absolute_error: 0.0786 - val_loss: 0.1259 - val_mean_absolute_error: 0.1865\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0175 - mean_absolute_error: 0.0842 - val_loss: 0.1307 - val_mean_absolute_error: 0.1725\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0174 - mean_absolute_error: 0.0857 - val_loss: 0.1263 - val_mean_absolute_error: 0.1932\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0162 - mean_absolute_error: 0.0828 - val_loss: 0.1141 - val_mean_absolute_error: 0.1582\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0103 - mean_absolute_error: 0.0711 - val_loss: 0.1437 - val_mean_absolute_error: 0.1842\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0088 - mean_absolute_error: 0.0653 - val_loss: 0.1299 - val_mean_absolute_error: 0.1675\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0100 - mean_absolute_error: 0.0708 - val_loss: 0.1230 - val_mean_absolute_error: 0.1700\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0071 - mean_absolute_error: 0.0536 - val_loss: 0.1269 - val_mean_absolute_error: 0.1774\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0113 - mean_absolute_error: 0.0733 - val_loss: 0.1217 - val_mean_absolute_error: 0.1519\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0082 - mean_absolute_error: 0.0614 - val_loss: 0.1392 - val_mean_absolute_error: 0.1876\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0109 - mean_absolute_error: 0.0771 - val_loss: 0.1232 - val_mean_absolute_error: 0.1546\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0107 - mean_absolute_error: 0.0747 - val_loss: 0.1321 - val_mean_absolute_error: 0.1986\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0105 - mean_absolute_error: 0.0740 - val_loss: 0.1220 - val_mean_absolute_error: 0.1549\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0149 - mean_absolute_error: 0.0859 - val_loss: 0.1410 - val_mean_absolute_error: 0.2015\n",
      "2/2 [==============================] - 0s 89ms/step - loss: 0.1410 - mean_absolute_error: 0.2015\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import random\n",
    "import sys\n",
    "import eli5\n",
    "\n",
    "import d2d\n",
    "import importlib\n",
    "importlib.reload(d2d)\n",
    "import pandas as pd\n",
    "import os\n",
    "\n",
    "SEED = 1\n",
    "random.seed(SEED)\n",
    "np.random.seed(SEED)\n",
    "tf.random.set_seed(SEED)\n",
    "\n",
    "file = \"/data/fast0/datasets/Rhone_data_continuous_highpass.h5\"\n",
    "\n",
    "filt = 'Highpass'\n",
    "\n",
    "history = {}\n",
    "val_performance = {}\n",
    "performance = {}\n",
    "\n",
    "ww = 200\n",
    "bs = 32\n",
    "\n",
    "random.seed(SEED)\n",
    "np.random.seed(SEED)\n",
    "tf.random.set_seed(SEED)\n",
    "\n",
    "input_columns = list(np.arange(0, 2308, 1))\n",
    "\n",
    "\n",
    "linear_model, lstm_model, dnn_model, df_all_chan, das_data_all, f  = d2d.import_data(filename = file, input_columns = input_columns)\n",
    "\n",
    "da = df_all_chan.to_numpy()\n",
    "\n",
    "multi_step_window_shuffled = d2d.WindowGenerator(df_all_chan,\n",
    "                                                input_width=ww,\n",
    "                                                label_width=1, \n",
    "                                                shift=0,\n",
    "                                                label_columns=['Discharge'],\n",
    "                                                input_columns=input_columns,\n",
    "                                                shuffle=True,\n",
    "                                                batch_size=bs)\n",
    "\n",
    "history['lstm_whole_cable'] = d2d.compile_and_fit(lstm_model, multi_step_window_shuffled, learning_rate = 0.001)\n",
    "\n",
    "val_performance['lstm_whole_cable'] = lstm_model.evaluate(multi_step_window_shuffled.val)\n",
    "performance['lstm_whole_cable'] = lstm_model.evaluate(multi_step_window_shuffled.test, verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "62b66a18",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "9f577c51",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2308\n"
     ]
    }
   ],
   "source": [
    "prediction_set = d2d.WindowGenerator(df_all_chan,\n",
    "                                    input_width=ww,\n",
    "                                    label_width=1, \n",
    "                                    shift=0,\n",
    "                                    label_columns=['Discharge'],\n",
    "                                    input_columns=input_columns,\n",
    "                                    shuffle=False,\n",
    "                                    batch_size=bs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "3d30c9f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = lstm_model.predict(prediction_set.train).squeeze()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "efce9fb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions_real = (predictions * prediction_set.dis_std) + prediction_set.dis_mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d72003e2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "f1d0d976",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "real_dis = []\n",
    "\n",
    "for i in prediction_set.train.as_numpy_iterator():\n",
    "    real_dis.append(i[-1])\n",
    "    \n",
    "real_dis = np.asarray(real_dis)\n",
    "real_dis = real_dis.reshape(real_dis.shape[0] * real_dis.shape[1])\n",
    "\n",
    "real_dis = (real_dis * prediction_set.dis_std) + prediction_set.dis_mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d9e375ca",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "6636b549",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n",
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "\n",
    "fig, ax = plt.subplots(figsize = [15,10])\n",
    "fig.patch.set_facecolor('white')\n",
    "\n",
    "ax.plot(predictions_real, label = 'Predicted', lw = 3, color = 'brown', linestyle = '--')\n",
    "ax.plot(real_dis, label = 'Observed', lw = 3, color = 'steelblue')\n",
    "\n",
    "ax.set_ylabel('Discharge (m$^3$/s)', size=20)\n",
    "ax.set_xlabel('Time Step', size=20)\n",
    "\n",
    "ax.tick_params(labelsize=16)\n",
    "\n",
    "ax.legend(prop={'size': 20})\n",
    "\n",
    "fig.savefig('discharge_train_timeseries.eps', dpi=1200)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "13238d04",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e55b9b4",
   "metadata": {},
   "source": [
    "# Bootstrapped residuals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "115755e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-55-509a33ba15ad>:19: DeprecationWarning: Seeding based on hashing is deprecated\n",
      "since Python 3.9 and will be removed in a subsequent version. The only \n",
      "supported seed types are: None, int, float, str, bytes, and bytearray.\n",
      "  random.seed(SEED)\n",
      "<ipython-input-55-509a33ba15ad>:34: DeprecationWarning: Seeding based on hashing is deprecated\n",
      "since Python 3.9 and will be removed in a subsequent version. The only \n",
      "supported seed types are: None, int, float, str, bytes, and bytearray.\n",
      "  random.seed(SEED)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 264ms/step - loss: 0.8842 - mean_absolute_error: 0.7757 - val_loss: 0.4469 - val_mean_absolute_error: 0.5254\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 200ms/step - loss: 0.5538 - mean_absolute_error: 0.5516 - val_loss: 0.5404 - val_mean_absolute_error: 0.6103\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 206ms/step - loss: 0.3882 - mean_absolute_error: 0.4912 - val_loss: 0.4451 - val_mean_absolute_error: 0.5200\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 224ms/step - loss: 0.3321 - mean_absolute_error: 0.4325 - val_loss: 0.3548 - val_mean_absolute_error: 0.4969\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 192ms/step - loss: 0.2493 - mean_absolute_error: 0.3749 - val_loss: 0.2874 - val_mean_absolute_error: 0.4320\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 213ms/step - loss: 0.1863 - mean_absolute_error: 0.2999 - val_loss: 0.2289 - val_mean_absolute_error: 0.3540\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 218ms/step - loss: 0.1418 - mean_absolute_error: 0.2636 - val_loss: 0.2229 - val_mean_absolute_error: 0.3269\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 203ms/step - loss: 0.1045 - mean_absolute_error: 0.2267 - val_loss: 0.1954 - val_mean_absolute_error: 0.3145\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 2s 250ms/step - loss: 0.0826 - mean_absolute_error: 0.1985 - val_loss: 0.2012 - val_mean_absolute_error: 0.3117\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.1050 - mean_absolute_error: 0.2170 - val_loss: 0.2317 - val_mean_absolute_error: 0.3621\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 224ms/step - loss: 0.1133 - mean_absolute_error: 0.2325 - val_loss: 0.1708 - val_mean_absolute_error: 0.2944\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 2s 385ms/step - loss: 0.0869 - mean_absolute_error: 0.2032 - val_loss: 0.1674 - val_mean_absolute_error: 0.2916\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.0617 - mean_absolute_error: 0.1713 - val_loss: 0.1350 - val_mean_absolute_error: 0.2531\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 220ms/step - loss: 0.0457 - mean_absolute_error: 0.1431 - val_loss: 0.1243 - val_mean_absolute_error: 0.2450\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 214ms/step - loss: 0.0318 - mean_absolute_error: 0.1240 - val_loss: 0.1146 - val_mean_absolute_error: 0.2134\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 219ms/step - loss: 0.0321 - mean_absolute_error: 0.1131 - val_loss: 0.1423 - val_mean_absolute_error: 0.2286\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 2s 234ms/step - loss: 0.0289 - mean_absolute_error: 0.1049 - val_loss: 0.1462 - val_mean_absolute_error: 0.2230\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 2s 253ms/step - loss: 0.0164 - mean_absolute_error: 0.0847 - val_loss: 0.1350 - val_mean_absolute_error: 0.2056\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 2s 276ms/step - loss: 0.0130 - mean_absolute_error: 0.0767 - val_loss: 0.1364 - val_mean_absolute_error: 0.2029\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 3s 409ms/step - loss: 0.0107 - mean_absolute_error: 0.0687 - val_loss: 0.1362 - val_mean_absolute_error: 0.1998\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 3s 481ms/step - loss: 0.0082 - mean_absolute_error: 0.0588 - val_loss: 0.1342 - val_mean_absolute_error: 0.1972\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 3s 448ms/step - loss: 0.0066 - mean_absolute_error: 0.0564 - val_loss: 0.1295 - val_mean_absolute_error: 0.1892\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 2s 355ms/step - loss: 0.0056 - mean_absolute_error: 0.0523 - val_loss: 0.1333 - val_mean_absolute_error: 0.1969\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 2s 284ms/step - loss: 0.0094 - mean_absolute_error: 0.0672 - val_loss: 0.1331 - val_mean_absolute_error: 0.1941\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 2s 265ms/step - loss: 0.0079 - mean_absolute_error: 0.0618 - val_loss: 0.1381 - val_mean_absolute_error: 0.2130\n",
      "2/2 [==============================] - 0s 112ms/step - loss: 0.1381 - mean_absolute_error: 0.2130\n",
      "2308\n",
      "Done with seed: 0\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 220ms/step - loss: 0.8887 - mean_absolute_error: 0.7776 - val_loss: 0.7887 - val_mean_absolute_error: 0.6859\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.5030 - mean_absolute_error: 0.5529 - val_loss: 0.5611 - val_mean_absolute_error: 0.5171\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.3162 - mean_absolute_error: 0.4405 - val_loss: 0.4273 - val_mean_absolute_error: 0.4555\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.2541 - mean_absolute_error: 0.3798 - val_loss: 0.4304 - val_mean_absolute_error: 0.4349\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.1770 - mean_absolute_error: 0.3030 - val_loss: 0.3465 - val_mean_absolute_error: 0.3993\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.1471 - mean_absolute_error: 0.2667 - val_loss: 0.2992 - val_mean_absolute_error: 0.3415\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.1122 - mean_absolute_error: 0.2350 - val_loss: 0.2728 - val_mean_absolute_error: 0.3184\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0812 - mean_absolute_error: 0.2023 - val_loss: 0.2383 - val_mean_absolute_error: 0.2976\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0685 - mean_absolute_error: 0.1819 - val_loss: 0.2161 - val_mean_absolute_error: 0.2694\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0568 - mean_absolute_error: 0.1609 - val_loss: 0.2127 - val_mean_absolute_error: 0.2484\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0405 - mean_absolute_error: 0.1370 - val_loss: 0.2084 - val_mean_absolute_error: 0.2396\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0355 - mean_absolute_error: 0.1215 - val_loss: 0.1882 - val_mean_absolute_error: 0.2311\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0267 - mean_absolute_error: 0.1054 - val_loss: 0.1805 - val_mean_absolute_error: 0.2213\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0271 - mean_absolute_error: 0.1067 - val_loss: 0.1771 - val_mean_absolute_error: 0.1942\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0234 - mean_absolute_error: 0.0974 - val_loss: 0.2299 - val_mean_absolute_error: 0.2442\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0326 - mean_absolute_error: 0.1217 - val_loss: 0.2170 - val_mean_absolute_error: 0.2161\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0297 - mean_absolute_error: 0.1119 - val_loss: 0.2152 - val_mean_absolute_error: 0.2731\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0394 - mean_absolute_error: 0.1375 - val_loss: 0.1828 - val_mean_absolute_error: 0.2081\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0265 - mean_absolute_error: 0.1095 - val_loss: 0.2153 - val_mean_absolute_error: 0.2402\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0364 - mean_absolute_error: 0.1312 - val_loss: 0.1995 - val_mean_absolute_error: 0.2367\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0241 - mean_absolute_error: 0.1027 - val_loss: 0.1805 - val_mean_absolute_error: 0.2090\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0305 - mean_absolute_error: 0.1200 - val_loss: 0.2006 - val_mean_absolute_error: 0.2502\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0261 - mean_absolute_error: 0.1195 - val_loss: 0.1879 - val_mean_absolute_error: 0.1956\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 133ms/step - loss: 0.0246 - mean_absolute_error: 0.1027 - val_loss: 0.1969 - val_mean_absolute_error: 0.2289\n",
      "2/2 [==============================] - 0s 64ms/step - loss: 0.1969 - mean_absolute_error: 0.2289\n",
      "2308\n",
      "Done with seed: 1\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 266ms/step - loss: 0.7782 - mean_absolute_error: 0.6986 - val_loss: 0.3251 - val_mean_absolute_error: 0.4476\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 253ms/step - loss: 0.5307 - mean_absolute_error: 0.5738 - val_loss: 0.3314 - val_mean_absolute_error: 0.4713\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 228ms/step - loss: 0.3381 - mean_absolute_error: 0.4487 - val_loss: 0.2636 - val_mean_absolute_error: 0.3764\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.2312 - mean_absolute_error: 0.3549 - val_loss: 0.2765 - val_mean_absolute_error: 0.3970\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.1788 - mean_absolute_error: 0.3059 - val_loss: 0.2468 - val_mean_absolute_error: 0.3539\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.1368 - mean_absolute_error: 0.2655 - val_loss: 0.2401 - val_mean_absolute_error: 0.3409\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 188ms/step - loss: 0.0975 - mean_absolute_error: 0.2293 - val_loss: 0.2013 - val_mean_absolute_error: 0.3033\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 188ms/step - loss: 0.0700 - mean_absolute_error: 0.1839 - val_loss: 0.1687 - val_mean_absolute_error: 0.2752\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 188ms/step - loss: 0.0531 - mean_absolute_error: 0.1637 - val_loss: 0.1578 - val_mean_absolute_error: 0.2534\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0377 - mean_absolute_error: 0.1337 - val_loss: 0.1669 - val_mean_absolute_error: 0.2449\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 188ms/step - loss: 0.0255 - mean_absolute_error: 0.1150 - val_loss: 0.1627 - val_mean_absolute_error: 0.2289\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0211 - mean_absolute_error: 0.1026 - val_loss: 0.1718 - val_mean_absolute_error: 0.2160\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 194ms/step - loss: 0.0203 - mean_absolute_error: 0.0933 - val_loss: 0.1820 - val_mean_absolute_error: 0.2322\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 192ms/step - loss: 0.0254 - mean_absolute_error: 0.1065 - val_loss: 0.1612 - val_mean_absolute_error: 0.2069\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 2s 234ms/step - loss: 0.0171 - mean_absolute_error: 0.0873 - val_loss: 0.1484 - val_mean_absolute_error: 0.2148\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 2s 246ms/step - loss: 0.0180 - mean_absolute_error: 0.0917 - val_loss: 0.1515 - val_mean_absolute_error: 0.2071\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 2s 267ms/step - loss: 0.0229 - mean_absolute_error: 0.1024 - val_loss: 0.1527 - val_mean_absolute_error: 0.2071\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 2s 265ms/step - loss: 0.0176 - mean_absolute_error: 0.0966 - val_loss: 0.1671 - val_mean_absolute_error: 0.2109\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 2s 238ms/step - loss: 0.0228 - mean_absolute_error: 0.1052 - val_loss: 0.1597 - val_mean_absolute_error: 0.2133\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 2s 227ms/step - loss: 0.0142 - mean_absolute_error: 0.0895 - val_loss: 0.1577 - val_mean_absolute_error: 0.2230\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0232 - mean_absolute_error: 0.1098 - val_loss: 0.1575 - val_mean_absolute_error: 0.2002\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0186 - mean_absolute_error: 0.0993 - val_loss: 0.1563 - val_mean_absolute_error: 0.1986\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0176 - mean_absolute_error: 0.0949 - val_loss: 0.1568 - val_mean_absolute_error: 0.2236\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0105 - mean_absolute_error: 0.0759 - val_loss: 0.1486 - val_mean_absolute_error: 0.1761\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0082 - mean_absolute_error: 0.0657 - val_loss: 0.1495 - val_mean_absolute_error: 0.1896\n",
      "2/2 [==============================] - 0s 88ms/step - loss: 0.1495 - mean_absolute_error: 0.1896\n",
      "2308\n",
      "Done with seed: 2\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 219ms/step - loss: 1.0815 - mean_absolute_error: 0.8381 - val_loss: 0.5262 - val_mean_absolute_error: 0.5928\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.5301 - mean_absolute_error: 0.5710 - val_loss: 0.4265 - val_mean_absolute_error: 0.5060\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.3421 - mean_absolute_error: 0.4501 - val_loss: 0.3598 - val_mean_absolute_error: 0.4753\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.2705 - mean_absolute_error: 0.3828 - val_loss: 0.2797 - val_mean_absolute_error: 0.4088\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.1973 - mean_absolute_error: 0.3170 - val_loss: 0.2417 - val_mean_absolute_error: 0.3681\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.1487 - mean_absolute_error: 0.2689 - val_loss: 0.2202 - val_mean_absolute_error: 0.3280\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.1244 - mean_absolute_error: 0.2413 - val_loss: 0.2112 - val_mean_absolute_error: 0.3115\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.1001 - mean_absolute_error: 0.2174 - val_loss: 0.2042 - val_mean_absolute_error: 0.3028\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0878 - mean_absolute_error: 0.1988 - val_loss: 0.1805 - val_mean_absolute_error: 0.2736\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0679 - mean_absolute_error: 0.1710 - val_loss: 0.1662 - val_mean_absolute_error: 0.2507\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0551 - mean_absolute_error: 0.1513 - val_loss: 0.1494 - val_mean_absolute_error: 0.2366\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0443 - mean_absolute_error: 0.1338 - val_loss: 0.1498 - val_mean_absolute_error: 0.2401\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0349 - mean_absolute_error: 0.1161 - val_loss: 0.1455 - val_mean_absolute_error: 0.2216\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0283 - mean_absolute_error: 0.1030 - val_loss: 0.1390 - val_mean_absolute_error: 0.2112\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0262 - mean_absolute_error: 0.1012 - val_loss: 0.1343 - val_mean_absolute_error: 0.2129\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0224 - mean_absolute_error: 0.0885 - val_loss: 0.1233 - val_mean_absolute_error: 0.1961\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0221 - mean_absolute_error: 0.0862 - val_loss: 0.1348 - val_mean_absolute_error: 0.1993\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0165 - mean_absolute_error: 0.0719 - val_loss: 0.1218 - val_mean_absolute_error: 0.1785\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0191 - mean_absolute_error: 0.0717 - val_loss: 0.1214 - val_mean_absolute_error: 0.1833\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0136 - mean_absolute_error: 0.0670 - val_loss: 0.1203 - val_mean_absolute_error: 0.1843\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0176 - mean_absolute_error: 0.0698 - val_loss: 0.1261 - val_mean_absolute_error: 0.1901\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0152 - mean_absolute_error: 0.0656 - val_loss: 0.1177 - val_mean_absolute_error: 0.1741\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0167 - mean_absolute_error: 0.0642 - val_loss: 0.1160 - val_mean_absolute_error: 0.1678\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0116 - mean_absolute_error: 0.0543 - val_loss: 0.1186 - val_mean_absolute_error: 0.1675\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0118 - mean_absolute_error: 0.0506 - val_loss: 0.1211 - val_mean_absolute_error: 0.1649\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0101 - mean_absolute_error: 0.0481 - val_loss: 0.1279 - val_mean_absolute_error: 0.1749\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0091 - mean_absolute_error: 0.0469 - val_loss: 0.1310 - val_mean_absolute_error: 0.1876\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0097 - mean_absolute_error: 0.0536 - val_loss: 0.1297 - val_mean_absolute_error: 0.1781\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0141 - mean_absolute_error: 0.0714 - val_loss: 0.1097 - val_mean_absolute_error: 0.1678\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0195 - mean_absolute_error: 0.0839 - val_loss: 0.1108 - val_mean_absolute_error: 0.1934\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0251 - mean_absolute_error: 0.0957 - val_loss: 0.1197 - val_mean_absolute_error: 0.1912\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0232 - mean_absolute_error: 0.1004 - val_loss: 0.1433 - val_mean_absolute_error: 0.2168\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0204 - mean_absolute_error: 0.0959 - val_loss: 0.1395 - val_mean_absolute_error: 0.2016\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0277 - mean_absolute_error: 0.1063 - val_loss: 0.1236 - val_mean_absolute_error: 0.2137\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0257 - mean_absolute_error: 0.1060 - val_loss: 0.1308 - val_mean_absolute_error: 0.1972\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0181 - mean_absolute_error: 0.0887 - val_loss: 0.1368 - val_mean_absolute_error: 0.2067\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0181 - mean_absolute_error: 0.0900 - val_loss: 0.1179 - val_mean_absolute_error: 0.1684\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0141 - mean_absolute_error: 0.0705 - val_loss: 0.1077 - val_mean_absolute_error: 0.1719\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 2s 245ms/step - loss: 0.0125 - mean_absolute_error: 0.0655 - val_loss: 0.1307 - val_mean_absolute_error: 0.1779\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 2s 231ms/step - loss: 0.0112 - mean_absolute_error: 0.0505 - val_loss: 0.1339 - val_mean_absolute_error: 0.1735\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 2s 233ms/step - loss: 0.0095 - mean_absolute_error: 0.0471 - val_loss: 0.1206 - val_mean_absolute_error: 0.1597\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 2s 236ms/step - loss: 0.0088 - mean_absolute_error: 0.0422 - val_loss: 0.1277 - val_mean_absolute_error: 0.1631\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0078 - mean_absolute_error: 0.0359 - val_loss: 0.1323 - val_mean_absolute_error: 0.1674\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 182ms/step - loss: 0.0079 - mean_absolute_error: 0.0380 - val_loss: 0.1205 - val_mean_absolute_error: 0.1526\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0073 - mean_absolute_error: 0.0313 - val_loss: 0.1202 - val_mean_absolute_error: 0.1526\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 182ms/step - loss: 0.0071 - mean_absolute_error: 0.0315 - val_loss: 0.1312 - val_mean_absolute_error: 0.1647\n",
      "Epoch 47/1000\n",
      "7/7 [==============================] - 1s 214ms/step - loss: 0.0070 - mean_absolute_error: 0.0313 - val_loss: 0.1274 - val_mean_absolute_error: 0.1558\n",
      "Epoch 48/1000\n",
      "7/7 [==============================] - 2s 291ms/step - loss: 0.0072 - mean_absolute_error: 0.0328 - val_loss: 0.1188 - val_mean_absolute_error: 0.1534\n",
      "2/2 [==============================] - 0s 93ms/step - loss: 0.1188 - mean_absolute_error: 0.1534\n",
      "2308\n",
      "Done with seed: 3\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 5s 462ms/step - loss: 0.8834 - mean_absolute_error: 0.7476 - val_loss: 0.4191 - val_mean_absolute_error: 0.5050\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 323ms/step - loss: 0.4531 - mean_absolute_error: 0.5087 - val_loss: 0.3690 - val_mean_absolute_error: 0.4920\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 266ms/step - loss: 0.3513 - mean_absolute_error: 0.4542 - val_loss: 0.3010 - val_mean_absolute_error: 0.3929\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 204ms/step - loss: 0.2576 - mean_absolute_error: 0.3753 - val_loss: 0.2487 - val_mean_absolute_error: 0.3847\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.1968 - mean_absolute_error: 0.3301 - val_loss: 0.2102 - val_mean_absolute_error: 0.3334\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 227ms/step - loss: 0.1485 - mean_absolute_error: 0.2723 - val_loss: 0.1610 - val_mean_absolute_error: 0.2833\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 2s 219ms/step - loss: 0.1194 - mean_absolute_error: 0.2447 - val_loss: 0.1747 - val_mean_absolute_error: 0.3019\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 219ms/step - loss: 0.1051 - mean_absolute_error: 0.2198 - val_loss: 0.1488 - val_mean_absolute_error: 0.2548\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 211ms/step - loss: 0.0673 - mean_absolute_error: 0.1832 - val_loss: 0.1541 - val_mean_absolute_error: 0.2531\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0578 - mean_absolute_error: 0.1675 - val_loss: 0.1284 - val_mean_absolute_error: 0.2270\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0420 - mean_absolute_error: 0.1385 - val_loss: 0.1297 - val_mean_absolute_error: 0.2223\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0314 - mean_absolute_error: 0.1143 - val_loss: 0.1428 - val_mean_absolute_error: 0.2361\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0304 - mean_absolute_error: 0.1133 - val_loss: 0.1231 - val_mean_absolute_error: 0.2201\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0231 - mean_absolute_error: 0.1005 - val_loss: 0.1186 - val_mean_absolute_error: 0.1992\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0210 - mean_absolute_error: 0.0934 - val_loss: 0.1456 - val_mean_absolute_error: 0.2337\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0192 - mean_absolute_error: 0.0941 - val_loss: 0.1254 - val_mean_absolute_error: 0.2060\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0170 - mean_absolute_error: 0.0871 - val_loss: 0.1262 - val_mean_absolute_error: 0.2291\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0171 - mean_absolute_error: 0.0850 - val_loss: 0.1388 - val_mean_absolute_error: 0.2235\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0185 - mean_absolute_error: 0.0875 - val_loss: 0.1236 - val_mean_absolute_error: 0.2002\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0162 - mean_absolute_error: 0.0822 - val_loss: 0.1115 - val_mean_absolute_error: 0.2072\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0157 - mean_absolute_error: 0.0817 - val_loss: 0.1309 - val_mean_absolute_error: 0.2184\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0257 - mean_absolute_error: 0.0982 - val_loss: 0.1180 - val_mean_absolute_error: 0.2099\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0197 - mean_absolute_error: 0.0911 - val_loss: 0.1237 - val_mean_absolute_error: 0.2220\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0193 - mean_absolute_error: 0.0929 - val_loss: 0.1496 - val_mean_absolute_error: 0.2380\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0181 - mean_absolute_error: 0.0903 - val_loss: 0.1343 - val_mean_absolute_error: 0.2165\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0184 - mean_absolute_error: 0.0884 - val_loss: 0.1355 - val_mean_absolute_error: 0.2297\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0184 - mean_absolute_error: 0.0933 - val_loss: 0.1599 - val_mean_absolute_error: 0.2602\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0161 - mean_absolute_error: 0.0896 - val_loss: 0.1274 - val_mean_absolute_error: 0.2210\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0130 - mean_absolute_error: 0.0737 - val_loss: 0.1307 - val_mean_absolute_error: 0.2024\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0110 - mean_absolute_error: 0.0615 - val_loss: 0.1417 - val_mean_absolute_error: 0.2212\n",
      "2/2 [==============================] - 0s 67ms/step - loss: 0.1417 - mean_absolute_error: 0.2212\n",
      "2308\n",
      "Done with seed: 4\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 254ms/step - loss: 0.8844 - mean_absolute_error: 0.7574 - val_loss: 0.6653 - val_mean_absolute_error: 0.6414\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.5714 - mean_absolute_error: 0.5790 - val_loss: 0.3293 - val_mean_absolute_error: 0.4144\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 224ms/step - loss: 0.2770 - mean_absolute_error: 0.4117 - val_loss: 0.2163 - val_mean_absolute_error: 0.3182\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 2s 237ms/step - loss: 0.1821 - mean_absolute_error: 0.3145 - val_loss: 0.2476 - val_mean_absolute_error: 0.3282\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 2s 236ms/step - loss: 0.1506 - mean_absolute_error: 0.2801 - val_loss: 0.2262 - val_mean_absolute_error: 0.3162\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 244ms/step - loss: 0.1049 - mean_absolute_error: 0.2287 - val_loss: 0.2014 - val_mean_absolute_error: 0.2794\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 2s 237ms/step - loss: 0.0742 - mean_absolute_error: 0.1877 - val_loss: 0.1860 - val_mean_absolute_error: 0.2568\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 260ms/step - loss: 0.0781 - mean_absolute_error: 0.1903 - val_loss: 0.1891 - val_mean_absolute_error: 0.2633\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 2s 261ms/step - loss: 0.0979 - mean_absolute_error: 0.2203 - val_loss: 0.1871 - val_mean_absolute_error: 0.2881\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 2s 248ms/step - loss: 0.0825 - mean_absolute_error: 0.1940 - val_loss: 0.1496 - val_mean_absolute_error: 0.2368\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 241ms/step - loss: 0.0663 - mean_absolute_error: 0.1766 - val_loss: 0.1372 - val_mean_absolute_error: 0.2315\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 2s 249ms/step - loss: 0.0414 - mean_absolute_error: 0.1353 - val_loss: 0.1155 - val_mean_absolute_error: 0.2141\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 2s 243ms/step - loss: 0.0308 - mean_absolute_error: 0.1133 - val_loss: 0.1147 - val_mean_absolute_error: 0.2000\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 2s 263ms/step - loss: 0.0350 - mean_absolute_error: 0.1216 - val_loss: 0.1160 - val_mean_absolute_error: 0.2017\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 2s 272ms/step - loss: 0.0338 - mean_absolute_error: 0.1224 - val_loss: 0.1251 - val_mean_absolute_error: 0.2060\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 3s 387ms/step - loss: 0.0240 - mean_absolute_error: 0.1050 - val_loss: 0.1195 - val_mean_absolute_error: 0.1836\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 4s 566ms/step - loss: 0.0176 - mean_absolute_error: 0.0889 - val_loss: 0.1160 - val_mean_absolute_error: 0.1834\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 3s 486ms/step - loss: 0.0144 - mean_absolute_error: 0.0773 - val_loss: 0.1071 - val_mean_absolute_error: 0.1631\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 3s 427ms/step - loss: 0.0156 - mean_absolute_error: 0.0832 - val_loss: 0.1177 - val_mean_absolute_error: 0.1962\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 2s 310ms/step - loss: 0.0212 - mean_absolute_error: 0.0937 - val_loss: 0.1300 - val_mean_absolute_error: 0.2016\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 2s 257ms/step - loss: 0.0271 - mean_absolute_error: 0.1025 - val_loss: 0.1504 - val_mean_absolute_error: 0.2108\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 217ms/step - loss: 0.0395 - mean_absolute_error: 0.1321 - val_loss: 0.1686 - val_mean_absolute_error: 0.2548\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0460 - mean_absolute_error: 0.1553 - val_loss: 0.1657 - val_mean_absolute_error: 0.2205\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.0287 - mean_absolute_error: 0.1242 - val_loss: 0.1540 - val_mean_absolute_error: 0.2032\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0177 - mean_absolute_error: 0.1010 - val_loss: 0.1569 - val_mean_absolute_error: 0.2095\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0126 - mean_absolute_error: 0.0813 - val_loss: 0.1545 - val_mean_absolute_error: 0.2075\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0119 - mean_absolute_error: 0.0796 - val_loss: 0.1381 - val_mean_absolute_error: 0.1755\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0073 - mean_absolute_error: 0.0627 - val_loss: 0.1335 - val_mean_absolute_error: 0.1726\n",
      "2/2 [==============================] - 0s 80ms/step - loss: 0.1335 - mean_absolute_error: 0.1726\n",
      "2308\n",
      "Done with seed: 5\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 221ms/step - loss: 0.7712 - mean_absolute_error: 0.6957 - val_loss: 0.6202 - val_mean_absolute_error: 0.6077\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.4066 - mean_absolute_error: 0.4943 - val_loss: 0.3309 - val_mean_absolute_error: 0.4608\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.3238 - mean_absolute_error: 0.4349 - val_loss: 0.2824 - val_mean_absolute_error: 0.4231\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.2364 - mean_absolute_error: 0.3618 - val_loss: 0.2311 - val_mean_absolute_error: 0.3528\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.1583 - mean_absolute_error: 0.2901 - val_loss: 0.2075 - val_mean_absolute_error: 0.3246\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.1251 - mean_absolute_error: 0.2494 - val_loss: 0.1831 - val_mean_absolute_error: 0.2860\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0928 - mean_absolute_error: 0.2107 - val_loss: 0.1720 - val_mean_absolute_error: 0.2603\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0662 - mean_absolute_error: 0.1704 - val_loss: 0.1644 - val_mean_absolute_error: 0.2506\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0598 - mean_absolute_error: 0.1500 - val_loss: 0.1498 - val_mean_absolute_error: 0.2391\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0549 - mean_absolute_error: 0.1463 - val_loss: 0.1271 - val_mean_absolute_error: 0.2120\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0342 - mean_absolute_error: 0.1172 - val_loss: 0.1275 - val_mean_absolute_error: 0.2201\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0268 - mean_absolute_error: 0.1033 - val_loss: 0.1231 - val_mean_absolute_error: 0.1987\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0221 - mean_absolute_error: 0.0907 - val_loss: 0.1191 - val_mean_absolute_error: 0.1934\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0182 - mean_absolute_error: 0.0825 - val_loss: 0.1224 - val_mean_absolute_error: 0.1872\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0206 - mean_absolute_error: 0.0939 - val_loss: 0.1310 - val_mean_absolute_error: 0.2212\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0259 - mean_absolute_error: 0.1039 - val_loss: 0.1262 - val_mean_absolute_error: 0.1922\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0167 - mean_absolute_error: 0.0904 - val_loss: 0.1313 - val_mean_absolute_error: 0.1977\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0166 - mean_absolute_error: 0.0872 - val_loss: 0.1374 - val_mean_absolute_error: 0.2037\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0222 - mean_absolute_error: 0.0933 - val_loss: 0.1411 - val_mean_absolute_error: 0.2153\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0181 - mean_absolute_error: 0.0925 - val_loss: 0.1314 - val_mean_absolute_error: 0.1832\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0163 - mean_absolute_error: 0.0837 - val_loss: 0.1439 - val_mean_absolute_error: 0.2333\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0195 - mean_absolute_error: 0.0917 - val_loss: 0.1363 - val_mean_absolute_error: 0.1951\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0163 - mean_absolute_error: 0.0878 - val_loss: 0.1426 - val_mean_absolute_error: 0.2322\n",
      "2/2 [==============================] - 0s 87ms/step - loss: 0.1426 - mean_absolute_error: 0.2322\n",
      "2308\n",
      "Done with seed: 6\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 332ms/step - loss: 0.7676 - mean_absolute_error: 0.6956 - val_loss: 0.5758 - val_mean_absolute_error: 0.5745\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 275ms/step - loss: 0.3794 - mean_absolute_error: 0.4911 - val_loss: 0.3298 - val_mean_absolute_error: 0.4528\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 318ms/step - loss: 0.2710 - mean_absolute_error: 0.4004 - val_loss: 0.3028 - val_mean_absolute_error: 0.3891\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 3s 372ms/step - loss: 0.1817 - mean_absolute_error: 0.3230 - val_loss: 0.2297 - val_mean_absolute_error: 0.3285\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 2s 310ms/step - loss: 0.1333 - mean_absolute_error: 0.2728 - val_loss: 0.2345 - val_mean_absolute_error: 0.3282\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 307ms/step - loss: 0.1071 - mean_absolute_error: 0.2390 - val_loss: 0.1837 - val_mean_absolute_error: 0.2789\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 201ms/step - loss: 0.0855 - mean_absolute_error: 0.2067 - val_loss: 0.1787 - val_mean_absolute_error: 0.2551\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 234ms/step - loss: 0.0636 - mean_absolute_error: 0.1792 - val_loss: 0.1600 - val_mean_absolute_error: 0.2322\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.0505 - mean_absolute_error: 0.1570 - val_loss: 0.1427 - val_mean_absolute_error: 0.2119\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0363 - mean_absolute_error: 0.1341 - val_loss: 0.1296 - val_mean_absolute_error: 0.2065\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0375 - mean_absolute_error: 0.1300 - val_loss: 0.1206 - val_mean_absolute_error: 0.1949\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.0370 - mean_absolute_error: 0.1340 - val_loss: 0.1196 - val_mean_absolute_error: 0.1942\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0281 - mean_absolute_error: 0.1175 - val_loss: 0.1092 - val_mean_absolute_error: 0.1808\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0202 - mean_absolute_error: 0.0973 - val_loss: 0.1023 - val_mean_absolute_error: 0.1616\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0133 - mean_absolute_error: 0.0782 - val_loss: 0.1032 - val_mean_absolute_error: 0.1605\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0105 - mean_absolute_error: 0.0667 - val_loss: 0.0992 - val_mean_absolute_error: 0.1562\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0099 - mean_absolute_error: 0.0636 - val_loss: 0.1018 - val_mean_absolute_error: 0.1515\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0085 - mean_absolute_error: 0.0596 - val_loss: 0.1052 - val_mean_absolute_error: 0.1647\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0058 - mean_absolute_error: 0.0507 - val_loss: 0.1035 - val_mean_absolute_error: 0.1617\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0092 - mean_absolute_error: 0.0623 - val_loss: 0.1035 - val_mean_absolute_error: 0.1562\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0086 - mean_absolute_error: 0.0618 - val_loss: 0.1255 - val_mean_absolute_error: 0.1932\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0150 - mean_absolute_error: 0.0831 - val_loss: 0.1086 - val_mean_absolute_error: 0.1602\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0163 - mean_absolute_error: 0.0887 - val_loss: 0.1157 - val_mean_absolute_error: 0.2126\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0164 - mean_absolute_error: 0.0924 - val_loss: 0.1025 - val_mean_absolute_error: 0.1583\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0174 - mean_absolute_error: 0.0920 - val_loss: 0.1366 - val_mean_absolute_error: 0.2031\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0176 - mean_absolute_error: 0.0947 - val_loss: 0.1068 - val_mean_absolute_error: 0.1603\n",
      "2/2 [==============================] - 0s 66ms/step - loss: 0.1068 - mean_absolute_error: 0.1603\n",
      "2308\n",
      "Done with seed: 7\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 221ms/step - loss: 1.2000 - mean_absolute_error: 0.8632 - val_loss: 0.6674 - val_mean_absolute_error: 0.7013\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 235ms/step - loss: 0.6134 - mean_absolute_error: 0.6337 - val_loss: 0.5297 - val_mean_absolute_error: 0.5455\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 217ms/step - loss: 0.3500 - mean_absolute_error: 0.4422 - val_loss: 0.4215 - val_mean_absolute_error: 0.5229\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 220ms/step - loss: 0.2880 - mean_absolute_error: 0.4014 - val_loss: 0.3333 - val_mean_absolute_error: 0.4708\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 2s 228ms/step - loss: 0.2114 - mean_absolute_error: 0.3435 - val_loss: 0.2825 - val_mean_absolute_error: 0.4261\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 222ms/step - loss: 0.1661 - mean_absolute_error: 0.2961 - val_loss: 0.2625 - val_mean_absolute_error: 0.4076\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 2s 241ms/step - loss: 0.1517 - mean_absolute_error: 0.2804 - val_loss: 0.2399 - val_mean_absolute_error: 0.3785\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 182ms/step - loss: 0.1161 - mean_absolute_error: 0.2389 - val_loss: 0.1960 - val_mean_absolute_error: 0.3137\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.0813 - mean_absolute_error: 0.2018 - val_loss: 0.1607 - val_mean_absolute_error: 0.2824\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0669 - mean_absolute_error: 0.1796 - val_loss: 0.1709 - val_mean_absolute_error: 0.2967\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 191ms/step - loss: 0.0618 - mean_absolute_error: 0.1740 - val_loss: 0.1580 - val_mean_absolute_error: 0.2944\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.0558 - mean_absolute_error: 0.1681 - val_loss: 0.1804 - val_mean_absolute_error: 0.3071\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.0577 - mean_absolute_error: 0.1624 - val_loss: 0.1544 - val_mean_absolute_error: 0.2797\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0510 - mean_absolute_error: 0.1557 - val_loss: 0.1370 - val_mean_absolute_error: 0.2507\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0353 - mean_absolute_error: 0.1266 - val_loss: 0.1398 - val_mean_absolute_error: 0.2527\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0274 - mean_absolute_error: 0.1151 - val_loss: 0.1346 - val_mean_absolute_error: 0.2457\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0218 - mean_absolute_error: 0.1043 - val_loss: 0.1268 - val_mean_absolute_error: 0.2274\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0174 - mean_absolute_error: 0.0877 - val_loss: 0.1345 - val_mean_absolute_error: 0.2348\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0163 - mean_absolute_error: 0.0832 - val_loss: 0.1296 - val_mean_absolute_error: 0.2120\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0149 - mean_absolute_error: 0.0772 - val_loss: 0.1349 - val_mean_absolute_error: 0.2392\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0128 - mean_absolute_error: 0.0722 - val_loss: 0.1235 - val_mean_absolute_error: 0.2180\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.0099 - mean_absolute_error: 0.0670 - val_loss: 0.1341 - val_mean_absolute_error: 0.2258\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0130 - mean_absolute_error: 0.0718 - val_loss: 0.1345 - val_mean_absolute_error: 0.2076\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 193ms/step - loss: 0.0090 - mean_absolute_error: 0.0617 - val_loss: 0.1317 - val_mean_absolute_error: 0.2276\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 202ms/step - loss: 0.0081 - mean_absolute_error: 0.0598 - val_loss: 0.1247 - val_mean_absolute_error: 0.1954\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 200ms/step - loss: 0.0048 - mean_absolute_error: 0.0437 - val_loss: 0.1284 - val_mean_absolute_error: 0.2042\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 2s 255ms/step - loss: 0.0054 - mean_absolute_error: 0.0459 - val_loss: 0.1288 - val_mean_absolute_error: 0.1975\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 2s 264ms/step - loss: 0.0033 - mean_absolute_error: 0.0352 - val_loss: 0.1256 - val_mean_absolute_error: 0.2006\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 2s 245ms/step - loss: 0.0028 - mean_absolute_error: 0.0327 - val_loss: 0.1315 - val_mean_absolute_error: 0.1924\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 2s 250ms/step - loss: 0.0036 - mean_absolute_error: 0.0363 - val_loss: 0.1339 - val_mean_absolute_error: 0.2055\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 2s 311ms/step - loss: 0.0034 - mean_absolute_error: 0.0363 - val_loss: 0.1322 - val_mean_absolute_error: 0.1974\n",
      "2/2 [==============================] - 0s 148ms/step - loss: 0.1322 - mean_absolute_error: 0.1974\n",
      "2308\n",
      "Done with seed: 8\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 7s 342ms/step - loss: 1.0621 - mean_absolute_error: 0.8352 - val_loss: 0.7227 - val_mean_absolute_error: 0.7296\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 293ms/step - loss: 0.5781 - mean_absolute_error: 0.6303 - val_loss: 0.4973 - val_mean_absolute_error: 0.5502\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 260ms/step - loss: 0.3732 - mean_absolute_error: 0.4905 - val_loss: 0.3984 - val_mean_absolute_error: 0.4705\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 2s 233ms/step - loss: 0.2819 - mean_absolute_error: 0.4139 - val_loss: 0.3513 - val_mean_absolute_error: 0.4428\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 213ms/step - loss: 0.2461 - mean_absolute_error: 0.3753 - val_loss: 0.2885 - val_mean_absolute_error: 0.3706\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 214ms/step - loss: 0.2280 - mean_absolute_error: 0.3554 - val_loss: 0.3060 - val_mean_absolute_error: 0.3900\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 205ms/step - loss: 0.2182 - mean_absolute_error: 0.3353 - val_loss: 0.2559 - val_mean_absolute_error: 0.3447\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.1668 - mean_absolute_error: 0.2931 - val_loss: 0.2244 - val_mean_absolute_error: 0.3112\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.1384 - mean_absolute_error: 0.2598 - val_loss: 0.1925 - val_mean_absolute_error: 0.2808\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 219ms/step - loss: 0.1132 - mean_absolute_error: 0.2256 - val_loss: 0.1857 - val_mean_absolute_error: 0.2572\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0892 - mean_absolute_error: 0.1988 - val_loss: 0.1675 - val_mean_absolute_error: 0.2399\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0717 - mean_absolute_error: 0.1722 - val_loss: 0.1718 - val_mean_absolute_error: 0.2384\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0619 - mean_absolute_error: 0.1555 - val_loss: 0.1597 - val_mean_absolute_error: 0.2201\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0496 - mean_absolute_error: 0.1365 - val_loss: 0.1348 - val_mean_absolute_error: 0.1973\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0395 - mean_absolute_error: 0.1215 - val_loss: 0.1283 - val_mean_absolute_error: 0.1840\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0308 - mean_absolute_error: 0.1065 - val_loss: 0.1263 - val_mean_absolute_error: 0.1816\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0251 - mean_absolute_error: 0.0941 - val_loss: 0.1271 - val_mean_absolute_error: 0.1797\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0185 - mean_absolute_error: 0.0802 - val_loss: 0.1159 - val_mean_absolute_error: 0.1579\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0172 - mean_absolute_error: 0.0730 - val_loss: 0.1163 - val_mean_absolute_error: 0.1627\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0196 - mean_absolute_error: 0.0813 - val_loss: 0.1430 - val_mean_absolute_error: 0.2006\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0301 - mean_absolute_error: 0.1119 - val_loss: 0.1450 - val_mean_absolute_error: 0.2027\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0363 - mean_absolute_error: 0.1209 - val_loss: 0.1215 - val_mean_absolute_error: 0.1945\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0231 - mean_absolute_error: 0.1055 - val_loss: 0.1213 - val_mean_absolute_error: 0.1741\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0174 - mean_absolute_error: 0.0857 - val_loss: 0.1408 - val_mean_absolute_error: 0.2139\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0210 - mean_absolute_error: 0.0976 - val_loss: 0.1244 - val_mean_absolute_error: 0.1738\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0130 - mean_absolute_error: 0.0764 - val_loss: 0.1139 - val_mean_absolute_error: 0.1679\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0116 - mean_absolute_error: 0.0717 - val_loss: 0.1265 - val_mean_absolute_error: 0.1834\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0115 - mean_absolute_error: 0.0721 - val_loss: 0.1127 - val_mean_absolute_error: 0.1484\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0083 - mean_absolute_error: 0.0606 - val_loss: 0.1123 - val_mean_absolute_error: 0.1590\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0083 - mean_absolute_error: 0.0598 - val_loss: 0.1265 - val_mean_absolute_error: 0.1619\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0067 - mean_absolute_error: 0.0512 - val_loss: 0.1222 - val_mean_absolute_error: 0.1655\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0069 - mean_absolute_error: 0.0535 - val_loss: 0.1017 - val_mean_absolute_error: 0.1472\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0071 - mean_absolute_error: 0.0508 - val_loss: 0.1105 - val_mean_absolute_error: 0.1550\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0146 - mean_absolute_error: 0.0693 - val_loss: 0.1419 - val_mean_absolute_error: 0.2091\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0192 - mean_absolute_error: 0.0844 - val_loss: 0.1430 - val_mean_absolute_error: 0.2032\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0159 - mean_absolute_error: 0.0866 - val_loss: 0.1062 - val_mean_absolute_error: 0.1466\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0095 - mean_absolute_error: 0.0650 - val_loss: 0.1057 - val_mean_absolute_error: 0.1388\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0071 - mean_absolute_error: 0.0522 - val_loss: 0.1215 - val_mean_absolute_error: 0.1635\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0067 - mean_absolute_error: 0.0486 - val_loss: 0.1094 - val_mean_absolute_error: 0.1385\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0044 - mean_absolute_error: 0.0436 - val_loss: 0.1022 - val_mean_absolute_error: 0.1384\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0040 - mean_absolute_error: 0.0402 - val_loss: 0.1003 - val_mean_absolute_error: 0.1376\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0036 - mean_absolute_error: 0.0379 - val_loss: 0.1014 - val_mean_absolute_error: 0.1426\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0035 - mean_absolute_error: 0.0375 - val_loss: 0.1074 - val_mean_absolute_error: 0.1341\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0037 - mean_absolute_error: 0.0384 - val_loss: 0.1098 - val_mean_absolute_error: 0.1357\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0039 - mean_absolute_error: 0.0426 - val_loss: 0.1047 - val_mean_absolute_error: 0.1494\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0040 - mean_absolute_error: 0.0410 - val_loss: 0.0994 - val_mean_absolute_error: 0.1420\n",
      "Epoch 47/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0038 - mean_absolute_error: 0.0428 - val_loss: 0.1015 - val_mean_absolute_error: 0.1424\n",
      "Epoch 48/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0038 - mean_absolute_error: 0.0419 - val_loss: 0.1020 - val_mean_absolute_error: 0.1278\n",
      "Epoch 49/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0036 - mean_absolute_error: 0.0423 - val_loss: 0.1134 - val_mean_absolute_error: 0.1622\n",
      "Epoch 50/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0046 - mean_absolute_error: 0.0472 - val_loss: 0.0999 - val_mean_absolute_error: 0.1319\n",
      "Epoch 51/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.0038 - mean_absolute_error: 0.0423 - val_loss: 0.1064 - val_mean_absolute_error: 0.1456\n",
      "Epoch 52/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0042 - mean_absolute_error: 0.0472 - val_loss: 0.1183 - val_mean_absolute_error: 0.1471\n",
      "Epoch 53/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0038 - mean_absolute_error: 0.0425 - val_loss: 0.1205 - val_mean_absolute_error: 0.1545\n",
      "Epoch 54/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0033 - mean_absolute_error: 0.0409 - val_loss: 0.1047 - val_mean_absolute_error: 0.1382\n",
      "Epoch 55/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0030 - mean_absolute_error: 0.0368 - val_loss: 0.1017 - val_mean_absolute_error: 0.1297\n",
      "Epoch 56/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0024 - mean_absolute_error: 0.0346 - val_loss: 0.1138 - val_mean_absolute_error: 0.1510\n",
      "2/2 [==============================] - 0s 72ms/step - loss: 0.1138 - mean_absolute_error: 0.1510\n",
      "2308\n",
      "Done with seed: 9\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 277ms/step - loss: 0.9422 - mean_absolute_error: 0.7510 - val_loss: 0.4903 - val_mean_absolute_error: 0.5287\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 259ms/step - loss: 0.4316 - mean_absolute_error: 0.5253 - val_loss: 0.4052 - val_mean_absolute_error: 0.5084\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 300ms/step - loss: 0.2892 - mean_absolute_error: 0.4218 - val_loss: 0.2826 - val_mean_absolute_error: 0.3793\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 3s 409ms/step - loss: 0.1920 - mean_absolute_error: 0.3390 - val_loss: 0.2429 - val_mean_absolute_error: 0.3504\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 2s 319ms/step - loss: 0.1570 - mean_absolute_error: 0.2927 - val_loss: 0.1960 - val_mean_absolute_error: 0.3182\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 292ms/step - loss: 0.1093 - mean_absolute_error: 0.2419 - val_loss: 0.1841 - val_mean_absolute_error: 0.2914\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 2s 277ms/step - loss: 0.0891 - mean_absolute_error: 0.2158 - val_loss: 0.1855 - val_mean_absolute_error: 0.2860\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 260ms/step - loss: 0.0741 - mean_absolute_error: 0.1971 - val_loss: 0.2123 - val_mean_absolute_error: 0.3120\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 206ms/step - loss: 0.0816 - mean_absolute_error: 0.2060 - val_loss: 0.1724 - val_mean_absolute_error: 0.2491\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 201ms/step - loss: 0.0571 - mean_absolute_error: 0.1624 - val_loss: 0.1799 - val_mean_absolute_error: 0.2603\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 206ms/step - loss: 0.0446 - mean_absolute_error: 0.1511 - val_loss: 0.1586 - val_mean_absolute_error: 0.2355\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0293 - mean_absolute_error: 0.1185 - val_loss: 0.1611 - val_mean_absolute_error: 0.2246\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 205ms/step - loss: 0.0267 - mean_absolute_error: 0.1129 - val_loss: 0.1838 - val_mean_absolute_error: 0.2155\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0215 - mean_absolute_error: 0.0964 - val_loss: 0.1756 - val_mean_absolute_error: 0.2161\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0205 - mean_absolute_error: 0.0947 - val_loss: 0.1704 - val_mean_absolute_error: 0.1960\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0150 - mean_absolute_error: 0.0721 - val_loss: 0.1720 - val_mean_absolute_error: 0.1960\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0148 - mean_absolute_error: 0.0713 - val_loss: 0.1761 - val_mean_absolute_error: 0.1882\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0157 - mean_absolute_error: 0.0728 - val_loss: 0.1897 - val_mean_absolute_error: 0.2112\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0177 - mean_absolute_error: 0.0797 - val_loss: 0.1789 - val_mean_absolute_error: 0.1966\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0199 - mean_absolute_error: 0.0885 - val_loss: 0.1828 - val_mean_absolute_error: 0.2231\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0197 - mean_absolute_error: 0.0889 - val_loss: 0.1834 - val_mean_absolute_error: 0.2180\n",
      "2/2 [==============================] - 0s 59ms/step - loss: 0.1834 - mean_absolute_error: 0.2180\n",
      "2308\n",
      "Done with seed: 10\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 210ms/step - loss: 0.7980 - mean_absolute_error: 0.7357 - val_loss: 0.4379 - val_mean_absolute_error: 0.5009\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.4064 - mean_absolute_error: 0.5002 - val_loss: 0.3201 - val_mean_absolute_error: 0.4380\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.2749 - mean_absolute_error: 0.3913 - val_loss: 0.2241 - val_mean_absolute_error: 0.3667\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.1847 - mean_absolute_error: 0.3247 - val_loss: 0.2330 - val_mean_absolute_error: 0.3555\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.1317 - mean_absolute_error: 0.2710 - val_loss: 0.2711 - val_mean_absolute_error: 0.3712\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.1102 - mean_absolute_error: 0.2501 - val_loss: 0.2907 - val_mean_absolute_error: 0.3726\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0989 - mean_absolute_error: 0.2316 - val_loss: 0.2543 - val_mean_absolute_error: 0.3424\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0638 - mean_absolute_error: 0.1784 - val_loss: 0.2289 - val_mean_absolute_error: 0.3058\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0519 - mean_absolute_error: 0.1615 - val_loss: 0.2101 - val_mean_absolute_error: 0.2735\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0372 - mean_absolute_error: 0.1355 - val_loss: 0.2241 - val_mean_absolute_error: 0.2901\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0349 - mean_absolute_error: 0.1261 - val_loss: 0.2121 - val_mean_absolute_error: 0.2588\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0220 - mean_absolute_error: 0.0988 - val_loss: 0.2249 - val_mean_absolute_error: 0.2850\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0281 - mean_absolute_error: 0.1156 - val_loss: 0.2100 - val_mean_absolute_error: 0.2674\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0172 - mean_absolute_error: 0.0970 - val_loss: 0.2199 - val_mean_absolute_error: 0.2662\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0151 - mean_absolute_error: 0.0883 - val_loss: 0.2226 - val_mean_absolute_error: 0.2435\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0113 - mean_absolute_error: 0.0744 - val_loss: 0.2167 - val_mean_absolute_error: 0.2534\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0092 - mean_absolute_error: 0.0672 - val_loss: 0.2138 - val_mean_absolute_error: 0.2399\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 135ms/step - loss: 0.0116 - mean_absolute_error: 0.0751 - val_loss: 0.2231 - val_mean_absolute_error: 0.2588\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0127 - mean_absolute_error: 0.0754 - val_loss: 0.2364 - val_mean_absolute_error: 0.2725\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0236 - mean_absolute_error: 0.0995 - val_loss: 0.2469 - val_mean_absolute_error: 0.3173\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0223 - mean_absolute_error: 0.1098 - val_loss: 0.1995 - val_mean_absolute_error: 0.2368\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0195 - mean_absolute_error: 0.1012 - val_loss: 0.2323 - val_mean_absolute_error: 0.2663\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0288 - mean_absolute_error: 0.1185 - val_loss: 0.2115 - val_mean_absolute_error: 0.2438\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0194 - mean_absolute_error: 0.1019 - val_loss: 0.2201 - val_mean_absolute_error: 0.2703\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0172 - mean_absolute_error: 0.0987 - val_loss: 0.2092 - val_mean_absolute_error: 0.2700\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0125 - mean_absolute_error: 0.0825 - val_loss: 0.2158 - val_mean_absolute_error: 0.2636\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0086 - mean_absolute_error: 0.0703 - val_loss: 0.1989 - val_mean_absolute_error: 0.2215\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0057 - mean_absolute_error: 0.0582 - val_loss: 0.1993 - val_mean_absolute_error: 0.2357\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0056 - mean_absolute_error: 0.0565 - val_loss: 0.2023 - val_mean_absolute_error: 0.2347\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0051 - mean_absolute_error: 0.0530 - val_loss: 0.1942 - val_mean_absolute_error: 0.2076\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0033 - mean_absolute_error: 0.0408 - val_loss: 0.1962 - val_mean_absolute_error: 0.2210\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0034 - mean_absolute_error: 0.0426 - val_loss: 0.2002 - val_mean_absolute_error: 0.2287\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0032 - mean_absolute_error: 0.0417 - val_loss: 0.1930 - val_mean_absolute_error: 0.2057\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0021 - mean_absolute_error: 0.0343 - val_loss: 0.1945 - val_mean_absolute_error: 0.2131\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0023 - mean_absolute_error: 0.0370 - val_loss: 0.1972 - val_mean_absolute_error: 0.2179\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0022 - mean_absolute_error: 0.0351 - val_loss: 0.1936 - val_mean_absolute_error: 0.2004\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0016 - mean_absolute_error: 0.0297 - val_loss: 0.1960 - val_mean_absolute_error: 0.2181\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0019 - mean_absolute_error: 0.0331 - val_loss: 0.1967 - val_mean_absolute_error: 0.2115\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0015 - mean_absolute_error: 0.0291 - val_loss: 0.1948 - val_mean_absolute_error: 0.2071\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0016 - mean_absolute_error: 0.0317 - val_loss: 0.1945 - val_mean_absolute_error: 0.2165\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0017 - mean_absolute_error: 0.0320 - val_loss: 0.1947 - val_mean_absolute_error: 0.2004\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0015 - mean_absolute_error: 0.0283 - val_loss: 0.1981 - val_mean_absolute_error: 0.2224\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0021 - mean_absolute_error: 0.0350 - val_loss: 0.1940 - val_mean_absolute_error: 0.2101\n",
      "2/2 [==============================] - 0s 91ms/step - loss: 0.1940 - mean_absolute_error: 0.2101\n",
      "2308\n",
      "Done with seed: 11\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 254ms/step - loss: 0.8180 - mean_absolute_error: 0.7486 - val_loss: 0.8005 - val_mean_absolute_error: 0.7053\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.4134 - mean_absolute_error: 0.4902 - val_loss: 0.5311 - val_mean_absolute_error: 0.5532\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.3127 - mean_absolute_error: 0.4230 - val_loss: 0.4764 - val_mean_absolute_error: 0.4999\n",
      "Epoch 4/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7/7 [==============================] - 1s 172ms/step - loss: 0.1841 - mean_absolute_error: 0.3249 - val_loss: 0.4141 - val_mean_absolute_error: 0.4759\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.1421 - mean_absolute_error: 0.2681 - val_loss: 0.3749 - val_mean_absolute_error: 0.4323\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0870 - mean_absolute_error: 0.2092 - val_loss: 0.3345 - val_mean_absolute_error: 0.3817\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0713 - mean_absolute_error: 0.1842 - val_loss: 0.3500 - val_mean_absolute_error: 0.3667\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0657 - mean_absolute_error: 0.1703 - val_loss: 0.3170 - val_mean_absolute_error: 0.3387\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0461 - mean_absolute_error: 0.1431 - val_loss: 0.3182 - val_mean_absolute_error: 0.3450\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0459 - mean_absolute_error: 0.1462 - val_loss: 0.3006 - val_mean_absolute_error: 0.3298\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0368 - mean_absolute_error: 0.1295 - val_loss: 0.2884 - val_mean_absolute_error: 0.3095\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0276 - mean_absolute_error: 0.1107 - val_loss: 0.2705 - val_mean_absolute_error: 0.2811\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0166 - mean_absolute_error: 0.0877 - val_loss: 0.2731 - val_mean_absolute_error: 0.2982\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0137 - mean_absolute_error: 0.0836 - val_loss: 0.2823 - val_mean_absolute_error: 0.2896\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0117 - mean_absolute_error: 0.0733 - val_loss: 0.3019 - val_mean_absolute_error: 0.2964\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 183ms/step - loss: 0.0155 - mean_absolute_error: 0.0836 - val_loss: 0.2976 - val_mean_absolute_error: 0.2953\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0240 - mean_absolute_error: 0.1064 - val_loss: 0.2971 - val_mean_absolute_error: 0.2976\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0181 - mean_absolute_error: 0.0976 - val_loss: 0.3243 - val_mean_absolute_error: 0.3419\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0310 - mean_absolute_error: 0.1357 - val_loss: 0.3102 - val_mean_absolute_error: 0.3306\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0264 - mean_absolute_error: 0.1173 - val_loss: 0.2998 - val_mean_absolute_error: 0.2747\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0180 - mean_absolute_error: 0.0978 - val_loss: 0.2985 - val_mean_absolute_error: 0.3154\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0171 - mean_absolute_error: 0.0957 - val_loss: 0.2736 - val_mean_absolute_error: 0.2909\n",
      "2/2 [==============================] - 0s 74ms/step - loss: 0.2736 - mean_absolute_error: 0.2909\n",
      "2308\n",
      "Done with seed: 12\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 237ms/step - loss: 1.1163 - mean_absolute_error: 0.8523 - val_loss: 0.5007 - val_mean_absolute_error: 0.5475\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 205ms/step - loss: 0.4391 - mean_absolute_error: 0.5087 - val_loss: 0.3563 - val_mean_absolute_error: 0.4533\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.2919 - mean_absolute_error: 0.4249 - val_loss: 0.2565 - val_mean_absolute_error: 0.3645\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.2166 - mean_absolute_error: 0.3547 - val_loss: 0.1958 - val_mean_absolute_error: 0.3212\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.1819 - mean_absolute_error: 0.3144 - val_loss: 0.1698 - val_mean_absolute_error: 0.3178\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 236ms/step - loss: 0.1520 - mean_absolute_error: 0.2872 - val_loss: 0.1285 - val_mean_absolute_error: 0.2780\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 2s 225ms/step - loss: 0.1089 - mean_absolute_error: 0.2419 - val_loss: 0.0912 - val_mean_absolute_error: 0.2278\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 230ms/step - loss: 0.0901 - mean_absolute_error: 0.2133 - val_loss: 0.0692 - val_mean_absolute_error: 0.1902\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 2s 228ms/step - loss: 0.0703 - mean_absolute_error: 0.1920 - val_loss: 0.0540 - val_mean_absolute_error: 0.1618\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 2s 233ms/step - loss: 0.0529 - mean_absolute_error: 0.1620 - val_loss: 0.0655 - val_mean_absolute_error: 0.1713\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.0434 - mean_absolute_error: 0.1441 - val_loss: 0.0498 - val_mean_absolute_error: 0.1586\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 218ms/step - loss: 0.0314 - mean_absolute_error: 0.1237 - val_loss: 0.0467 - val_mean_absolute_error: 0.1505\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 224ms/step - loss: 0.0236 - mean_absolute_error: 0.1049 - val_loss: 0.0355 - val_mean_absolute_error: 0.1198\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 216ms/step - loss: 0.0211 - mean_absolute_error: 0.0939 - val_loss: 0.0379 - val_mean_absolute_error: 0.1195\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 2s 225ms/step - loss: 0.0138 - mean_absolute_error: 0.0753 - val_loss: 0.0359 - val_mean_absolute_error: 0.1047\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 2s 227ms/step - loss: 0.0105 - mean_absolute_error: 0.0640 - val_loss: 0.0340 - val_mean_absolute_error: 0.0978\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 2s 236ms/step - loss: 0.0149 - mean_absolute_error: 0.0734 - val_loss: 0.0554 - val_mean_absolute_error: 0.1405\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 2s 247ms/step - loss: 0.0314 - mean_absolute_error: 0.1150 - val_loss: 0.0374 - val_mean_absolute_error: 0.1214\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 2s 234ms/step - loss: 0.0264 - mean_absolute_error: 0.1029 - val_loss: 0.0479 - val_mean_absolute_error: 0.1270\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 2s 248ms/step - loss: 0.0198 - mean_absolute_error: 0.0926 - val_loss: 0.0486 - val_mean_absolute_error: 0.1369\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 2s 283ms/step - loss: 0.0207 - mean_absolute_error: 0.0896 - val_loss: 0.0340 - val_mean_absolute_error: 0.1062\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 2s 276ms/step - loss: 0.0186 - mean_absolute_error: 0.0896 - val_loss: 0.0402 - val_mean_absolute_error: 0.1205\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 2s 251ms/step - loss: 0.0207 - mean_absolute_error: 0.0925 - val_loss: 0.0344 - val_mean_absolute_error: 0.1122\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 2s 242ms/step - loss: 0.0168 - mean_absolute_error: 0.0826 - val_loss: 0.0319 - val_mean_absolute_error: 0.0968\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 194ms/step - loss: 0.0171 - mean_absolute_error: 0.0678 - val_loss: 0.0296 - val_mean_absolute_error: 0.0913\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 183ms/step - loss: 0.0149 - mean_absolute_error: 0.0714 - val_loss: 0.0297 - val_mean_absolute_error: 0.0903\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 196ms/step - loss: 0.0121 - mean_absolute_error: 0.0625 - val_loss: 0.0339 - val_mean_absolute_error: 0.1014\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0114 - mean_absolute_error: 0.0632 - val_loss: 0.0303 - val_mean_absolute_error: 0.0842\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0123 - mean_absolute_error: 0.0578 - val_loss: 0.0362 - val_mean_absolute_error: 0.1054\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0131 - mean_absolute_error: 0.0746 - val_loss: 0.0309 - val_mean_absolute_error: 0.0937\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0094 - mean_absolute_error: 0.0704 - val_loss: 0.0403 - val_mean_absolute_error: 0.1215\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0128 - mean_absolute_error: 0.0819 - val_loss: 0.0369 - val_mean_absolute_error: 0.1004\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0092 - mean_absolute_error: 0.0621 - val_loss: 0.0319 - val_mean_absolute_error: 0.1081\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0094 - mean_absolute_error: 0.0709 - val_loss: 0.0304 - val_mean_absolute_error: 0.0800\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0074 - mean_absolute_error: 0.0619 - val_loss: 0.0352 - val_mean_absolute_error: 0.0998\n",
      "2/2 [==============================] - 0s 67ms/step - loss: 0.0352 - mean_absolute_error: 0.0998\n",
      "2308\n",
      "Done with seed: 13\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 219ms/step - loss: 0.8096 - mean_absolute_error: 0.7063 - val_loss: 0.4461 - val_mean_absolute_error: 0.5116\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.4453 - mean_absolute_error: 0.5191 - val_loss: 0.3511 - val_mean_absolute_error: 0.4725\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.2657 - mean_absolute_error: 0.4020 - val_loss: 0.3495 - val_mean_absolute_error: 0.4520\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.2114 - mean_absolute_error: 0.3412 - val_loss: 0.3253 - val_mean_absolute_error: 0.3991\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.1625 - mean_absolute_error: 0.2850 - val_loss: 0.2770 - val_mean_absolute_error: 0.3457\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.1105 - mean_absolute_error: 0.2309 - val_loss: 0.2320 - val_mean_absolute_error: 0.3318\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0876 - mean_absolute_error: 0.2061 - val_loss: 0.2140 - val_mean_absolute_error: 0.3144\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0673 - mean_absolute_error: 0.1800 - val_loss: 0.2128 - val_mean_absolute_error: 0.3083\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0512 - mean_absolute_error: 0.1611 - val_loss: 0.1846 - val_mean_absolute_error: 0.2729\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0342 - mean_absolute_error: 0.1334 - val_loss: 0.1857 - val_mean_absolute_error: 0.2767\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0271 - mean_absolute_error: 0.1089 - val_loss: 0.1918 - val_mean_absolute_error: 0.2772\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0191 - mean_absolute_error: 0.0890 - val_loss: 0.1770 - val_mean_absolute_error: 0.2601\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0156 - mean_absolute_error: 0.0813 - val_loss: 0.1721 - val_mean_absolute_error: 0.2532\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0159 - mean_absolute_error: 0.0813 - val_loss: 0.1640 - val_mean_absolute_error: 0.2467\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0140 - mean_absolute_error: 0.0813 - val_loss: 0.1894 - val_mean_absolute_error: 0.2718\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0127 - mean_absolute_error: 0.0811 - val_loss: 0.1958 - val_mean_absolute_error: 0.2861\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0204 - mean_absolute_error: 0.0955 - val_loss: 0.1872 - val_mean_absolute_error: 0.2545\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0361 - mean_absolute_error: 0.1239 - val_loss: 0.2113 - val_mean_absolute_error: 0.2964\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0291 - mean_absolute_error: 0.1194 - val_loss: 0.2020 - val_mean_absolute_error: 0.2906\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0328 - mean_absolute_error: 0.1231 - val_loss: 0.1952 - val_mean_absolute_error: 0.2964\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0259 - mean_absolute_error: 0.1195 - val_loss: 0.1965 - val_mean_absolute_error: 0.2616\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0166 - mean_absolute_error: 0.0938 - val_loss: 0.2184 - val_mean_absolute_error: 0.2954\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0149 - mean_absolute_error: 0.0869 - val_loss: 0.1872 - val_mean_absolute_error: 0.2562\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0084 - mean_absolute_error: 0.0667 - val_loss: 0.1848 - val_mean_absolute_error: 0.2441\n",
      "2/2 [==============================] - 0s 64ms/step - loss: 0.1848 - mean_absolute_error: 0.2441\n",
      "2308\n",
      "Done with seed: 14\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 246ms/step - loss: 0.9891 - mean_absolute_error: 0.8062 - val_loss: 0.5856 - val_mean_absolute_error: 0.6253\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.5311 - mean_absolute_error: 0.5823 - val_loss: 0.4003 - val_mean_absolute_error: 0.5094\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.3152 - mean_absolute_error: 0.4491 - val_loss: 0.3392 - val_mean_absolute_error: 0.4819\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 207ms/step - loss: 0.2214 - mean_absolute_error: 0.3659 - val_loss: 0.2252 - val_mean_absolute_error: 0.3784\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.1599 - mean_absolute_error: 0.2946 - val_loss: 0.1994 - val_mean_absolute_error: 0.3574\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 196ms/step - loss: 0.1314 - mean_absolute_error: 0.2584 - val_loss: 0.1705 - val_mean_absolute_error: 0.3253\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0935 - mean_absolute_error: 0.2182 - val_loss: 0.1517 - val_mean_absolute_error: 0.2986\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 211ms/step - loss: 0.0725 - mean_absolute_error: 0.1870 - val_loss: 0.1207 - val_mean_absolute_error: 0.2607\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 2s 225ms/step - loss: 0.0552 - mean_absolute_error: 0.1628 - val_loss: 0.1011 - val_mean_absolute_error: 0.2343\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 2s 239ms/step - loss: 0.0422 - mean_absolute_error: 0.1412 - val_loss: 0.1091 - val_mean_absolute_error: 0.2328\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 244ms/step - loss: 0.0399 - mean_absolute_error: 0.1287 - val_loss: 0.0896 - val_mean_absolute_error: 0.2071\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 3s 398ms/step - loss: 0.0359 - mean_absolute_error: 0.1209 - val_loss: 0.1099 - val_mean_absolute_error: 0.2415\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 3s 493ms/step - loss: 0.0398 - mean_absolute_error: 0.1342 - val_loss: 0.1140 - val_mean_absolute_error: 0.2469\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 3s 479ms/step - loss: 0.0394 - mean_absolute_error: 0.1271 - val_loss: 0.0876 - val_mean_absolute_error: 0.2020\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 3s 366ms/step - loss: 0.0247 - mean_absolute_error: 0.1009 - val_loss: 0.0872 - val_mean_absolute_error: 0.1966\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 2s 337ms/step - loss: 0.0255 - mean_absolute_error: 0.1006 - val_loss: 0.0989 - val_mean_absolute_error: 0.2109\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 2s 278ms/step - loss: 0.0216 - mean_absolute_error: 0.0934 - val_loss: 0.0792 - val_mean_absolute_error: 0.1718\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 199ms/step - loss: 0.0151 - mean_absolute_error: 0.0768 - val_loss: 0.0753 - val_mean_absolute_error: 0.1778\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0155 - mean_absolute_error: 0.0788 - val_loss: 0.0987 - val_mean_absolute_error: 0.2090\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0167 - mean_absolute_error: 0.0824 - val_loss: 0.0813 - val_mean_absolute_error: 0.1727\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0147 - mean_absolute_error: 0.0792 - val_loss: 0.0745 - val_mean_absolute_error: 0.1776\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0173 - mean_absolute_error: 0.0898 - val_loss: 0.1107 - val_mean_absolute_error: 0.2131\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0135 - mean_absolute_error: 0.0756 - val_loss: 0.0736 - val_mean_absolute_error: 0.1597\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0116 - mean_absolute_error: 0.0704 - val_loss: 0.0702 - val_mean_absolute_error: 0.1609\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0143 - mean_absolute_error: 0.0737 - val_loss: 0.1059 - val_mean_absolute_error: 0.2063\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0140 - mean_absolute_error: 0.0795 - val_loss: 0.0807 - val_mean_absolute_error: 0.1567\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0095 - mean_absolute_error: 0.0643 - val_loss: 0.0757 - val_mean_absolute_error: 0.1664\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0111 - mean_absolute_error: 0.0717 - val_loss: 0.1078 - val_mean_absolute_error: 0.2031\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0106 - mean_absolute_error: 0.0706 - val_loss: 0.0816 - val_mean_absolute_error: 0.1518\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0078 - mean_absolute_error: 0.0583 - val_loss: 0.0705 - val_mean_absolute_error: 0.1597\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0090 - mean_absolute_error: 0.0624 - val_loss: 0.1020 - val_mean_absolute_error: 0.1950\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0093 - mean_absolute_error: 0.0654 - val_loss: 0.0840 - val_mean_absolute_error: 0.1533\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0075 - mean_absolute_error: 0.0572 - val_loss: 0.0769 - val_mean_absolute_error: 0.1648\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0086 - mean_absolute_error: 0.0629 - val_loss: 0.1070 - val_mean_absolute_error: 0.1993\n",
      "2/2 [==============================] - 0s 69ms/step - loss: 0.1070 - mean_absolute_error: 0.1993\n",
      "2308\n",
      "Done with seed: 15\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 229ms/step - loss: 0.9274 - mean_absolute_error: 0.7781 - val_loss: 0.4738 - val_mean_absolute_error: 0.5350\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.4718 - mean_absolute_error: 0.5384 - val_loss: 0.3251 - val_mean_absolute_error: 0.4526\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 201ms/step - loss: 0.3642 - mean_absolute_error: 0.4520 - val_loss: 0.2246 - val_mean_absolute_error: 0.3881\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 211ms/step - loss: 0.2516 - mean_absolute_error: 0.3698 - val_loss: 0.1640 - val_mean_absolute_error: 0.3030\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 212ms/step - loss: 0.2146 - mean_absolute_error: 0.3338 - val_loss: 0.1523 - val_mean_absolute_error: 0.3053\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.1622 - mean_absolute_error: 0.2913 - val_loss: 0.1263 - val_mean_absolute_error: 0.2648\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 211ms/step - loss: 0.1385 - mean_absolute_error: 0.2606 - val_loss: 0.0865 - val_mean_absolute_error: 0.2106\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 191ms/step - loss: 0.1003 - mean_absolute_error: 0.2158 - val_loss: 0.0964 - val_mean_absolute_error: 0.2017\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.0760 - mean_absolute_error: 0.1902 - val_loss: 0.1056 - val_mean_absolute_error: 0.2110\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 203ms/step - loss: 0.0594 - mean_absolute_error: 0.1647 - val_loss: 0.0981 - val_mean_absolute_error: 0.1934\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 194ms/step - loss: 0.0517 - mean_absolute_error: 0.1516 - val_loss: 0.0892 - val_mean_absolute_error: 0.1959\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 196ms/step - loss: 0.0504 - mean_absolute_error: 0.1496 - val_loss: 0.0870 - val_mean_absolute_error: 0.1811\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 196ms/step - loss: 0.0484 - mean_absolute_error: 0.1489 - val_loss: 0.0943 - val_mean_absolute_error: 0.2151\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0338 - mean_absolute_error: 0.1279 - val_loss: 0.0805 - val_mean_absolute_error: 0.1861\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0305 - mean_absolute_error: 0.1176 - val_loss: 0.0702 - val_mean_absolute_error: 0.1572\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0215 - mean_absolute_error: 0.0969 - val_loss: 0.0782 - val_mean_absolute_error: 0.1570\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.0166 - mean_absolute_error: 0.0826 - val_loss: 0.0711 - val_mean_absolute_error: 0.1548\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0154 - mean_absolute_error: 0.0784 - val_loss: 0.0725 - val_mean_absolute_error: 0.1485\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0118 - mean_absolute_error: 0.0665 - val_loss: 0.0782 - val_mean_absolute_error: 0.1468\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0107 - mean_absolute_error: 0.0642 - val_loss: 0.0846 - val_mean_absolute_error: 0.1548\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0098 - mean_absolute_error: 0.0649 - val_loss: 0.0869 - val_mean_absolute_error: 0.1636\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.0083 - mean_absolute_error: 0.0597 - val_loss: 0.0831 - val_mean_absolute_error: 0.1453\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0167 - mean_absolute_error: 0.0775 - val_loss: 0.0766 - val_mean_absolute_error: 0.1387\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0158 - mean_absolute_error: 0.0841 - val_loss: 0.0930 - val_mean_absolute_error: 0.1689\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 193ms/step - loss: 0.0182 - mean_absolute_error: 0.0963 - val_loss: 0.0853 - val_mean_absolute_error: 0.1444\n",
      "2/2 [==============================] - 0s 86ms/step - loss: 0.0853 - mean_absolute_error: 0.1444\n",
      "2308\n",
      "Done with seed: 16\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 2s 185ms/step - loss: 0.8973 - mean_absolute_error: 0.7496 - val_loss: 0.4304 - val_mean_absolute_error: 0.5247\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.4356 - mean_absolute_error: 0.5141 - val_loss: 0.3553 - val_mean_absolute_error: 0.4511\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.3017 - mean_absolute_error: 0.4244 - val_loss: 0.2943 - val_mean_absolute_error: 0.3784\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.2294 - mean_absolute_error: 0.3548 - val_loss: 0.2764 - val_mean_absolute_error: 0.3664\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.1831 - mean_absolute_error: 0.3184 - val_loss: 0.1986 - val_mean_absolute_error: 0.2943\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.1240 - mean_absolute_error: 0.2529 - val_loss: 0.2003 - val_mean_absolute_error: 0.2861\n",
      "Epoch 7/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0912 - mean_absolute_error: 0.2161 - val_loss: 0.2069 - val_mean_absolute_error: 0.2783\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.0696 - mean_absolute_error: 0.1809 - val_loss: 0.1937 - val_mean_absolute_error: 0.2482\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0519 - mean_absolute_error: 0.1539 - val_loss: 0.2077 - val_mean_absolute_error: 0.2381\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0419 - mean_absolute_error: 0.1353 - val_loss: 0.2095 - val_mean_absolute_error: 0.2333\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 133ms/step - loss: 0.0334 - mean_absolute_error: 0.1237 - val_loss: 0.1915 - val_mean_absolute_error: 0.2094\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0289 - mean_absolute_error: 0.1120 - val_loss: 0.1792 - val_mean_absolute_error: 0.2026\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0223 - mean_absolute_error: 0.1003 - val_loss: 0.1742 - val_mean_absolute_error: 0.1958\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0191 - mean_absolute_error: 0.0886 - val_loss: 0.1692 - val_mean_absolute_error: 0.1881\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0127 - mean_absolute_error: 0.0698 - val_loss: 0.1631 - val_mean_absolute_error: 0.1803\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0093 - mean_absolute_error: 0.0562 - val_loss: 0.1607 - val_mean_absolute_error: 0.1703\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0069 - mean_absolute_error: 0.0468 - val_loss: 0.1633 - val_mean_absolute_error: 0.1689\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0060 - mean_absolute_error: 0.0419 - val_loss: 0.1630 - val_mean_absolute_error: 0.1643\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0048 - mean_absolute_error: 0.0367 - val_loss: 0.1651 - val_mean_absolute_error: 0.1665\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0046 - mean_absolute_error: 0.0364 - val_loss: 0.1634 - val_mean_absolute_error: 0.1644\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0056 - mean_absolute_error: 0.0431 - val_loss: 0.1619 - val_mean_absolute_error: 0.1637\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0079 - mean_absolute_error: 0.0519 - val_loss: 0.1613 - val_mean_absolute_error: 0.1606\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0113 - mean_absolute_error: 0.0670 - val_loss: 0.1624 - val_mean_absolute_error: 0.1881\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0134 - mean_absolute_error: 0.0781 - val_loss: 0.1580 - val_mean_absolute_error: 0.2010\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0205 - mean_absolute_error: 0.0974 - val_loss: 0.1466 - val_mean_absolute_error: 0.1723\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0334 - mean_absolute_error: 0.1332 - val_loss: 0.2193 - val_mean_absolute_error: 0.2776\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0560 - mean_absolute_error: 0.1736 - val_loss: 0.1853 - val_mean_absolute_error: 0.2141\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0433 - mean_absolute_error: 0.1561 - val_loss: 0.1667 - val_mean_absolute_error: 0.2151\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0327 - mean_absolute_error: 0.1313 - val_loss: 0.1823 - val_mean_absolute_error: 0.2071\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0193 - mean_absolute_error: 0.1001 - val_loss: 0.1805 - val_mean_absolute_error: 0.1987\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0124 - mean_absolute_error: 0.0817 - val_loss: 0.1638 - val_mean_absolute_error: 0.1814\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0089 - mean_absolute_error: 0.0660 - val_loss: 0.1575 - val_mean_absolute_error: 0.1709\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0059 - mean_absolute_error: 0.0521 - val_loss: 0.1501 - val_mean_absolute_error: 0.1769\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0076 - mean_absolute_error: 0.0588 - val_loss: 0.1535 - val_mean_absolute_error: 0.1634\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0083 - mean_absolute_error: 0.0588 - val_loss: 0.1673 - val_mean_absolute_error: 0.1892\n",
      "2/2 [==============================] - 0s 66ms/step - loss: 0.1673 - mean_absolute_error: 0.1892\n",
      "2308\n",
      "Done with seed: 17\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 247ms/step - loss: 0.9339 - mean_absolute_error: 0.7618 - val_loss: 0.5305 - val_mean_absolute_error: 0.6041\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 226ms/step - loss: 0.5154 - mean_absolute_error: 0.5912 - val_loss: 0.4017 - val_mean_absolute_error: 0.4867\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.3331 - mean_absolute_error: 0.4447 - val_loss: 0.3545 - val_mean_absolute_error: 0.4652\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.2401 - mean_absolute_error: 0.3814 - val_loss: 0.3572 - val_mean_absolute_error: 0.4088\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.1859 - mean_absolute_error: 0.3116 - val_loss: 0.3278 - val_mean_absolute_error: 0.4104\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.1448 - mean_absolute_error: 0.2841 - val_loss: 0.3222 - val_mean_absolute_error: 0.3968\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.1204 - mean_absolute_error: 0.2519 - val_loss: 0.2906 - val_mean_absolute_error: 0.3616\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0932 - mean_absolute_error: 0.2194 - val_loss: 0.2731 - val_mean_absolute_error: 0.3390\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 199ms/step - loss: 0.0756 - mean_absolute_error: 0.1900 - val_loss: 0.2771 - val_mean_absolute_error: 0.3208\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 210ms/step - loss: 0.0698 - mean_absolute_error: 0.1661 - val_loss: 0.2760 - val_mean_absolute_error: 0.3166\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 226ms/step - loss: 0.0681 - mean_absolute_error: 0.1559 - val_loss: 0.3038 - val_mean_absolute_error: 0.3348\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 219ms/step - loss: 0.1090 - mean_absolute_error: 0.1750 - val_loss: 0.2901 - val_mean_absolute_error: 0.3121\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0972 - mean_absolute_error: 0.1732 - val_loss: 0.2717 - val_mean_absolute_error: 0.3162\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 197ms/step - loss: 0.0718 - mean_absolute_error: 0.1629 - val_loss: 0.2789 - val_mean_absolute_error: 0.3154\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0587 - mean_absolute_error: 0.1466 - val_loss: 0.2699 - val_mean_absolute_error: 0.3146\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 192ms/step - loss: 0.0497 - mean_absolute_error: 0.1307 - val_loss: 0.2593 - val_mean_absolute_error: 0.2976\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0494 - mean_absolute_error: 0.1396 - val_loss: 0.2957 - val_mean_absolute_error: 0.3386\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0526 - mean_absolute_error: 0.1556 - val_loss: 0.3892 - val_mean_absolute_error: 0.4037\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0739 - mean_absolute_error: 0.1938 - val_loss: 0.2718 - val_mean_absolute_error: 0.2965\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0605 - mean_absolute_error: 0.1721 - val_loss: 0.3032 - val_mean_absolute_error: 0.3657\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0654 - mean_absolute_error: 0.1828 - val_loss: 0.2551 - val_mean_absolute_error: 0.2987\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0404 - mean_absolute_error: 0.1495 - val_loss: 0.2290 - val_mean_absolute_error: 0.2959\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0334 - mean_absolute_error: 0.1329 - val_loss: 0.2161 - val_mean_absolute_error: 0.2722\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0233 - mean_absolute_error: 0.1092 - val_loss: 0.2163 - val_mean_absolute_error: 0.2837\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0171 - mean_absolute_error: 0.0889 - val_loss: 0.2302 - val_mean_absolute_error: 0.2726\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0161 - mean_absolute_error: 0.0867 - val_loss: 0.2419 - val_mean_absolute_error: 0.2922\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.0176 - mean_absolute_error: 0.0939 - val_loss: 0.1996 - val_mean_absolute_error: 0.2440\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0154 - mean_absolute_error: 0.0853 - val_loss: 0.2184 - val_mean_absolute_error: 0.2929\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0186 - mean_absolute_error: 0.0958 - val_loss: 0.2264 - val_mean_absolute_error: 0.2646\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0134 - mean_absolute_error: 0.0808 - val_loss: 0.2217 - val_mean_absolute_error: 0.2705\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0133 - mean_absolute_error: 0.0819 - val_loss: 0.1955 - val_mean_absolute_error: 0.2586\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0119 - mean_absolute_error: 0.0781 - val_loss: 0.1996 - val_mean_absolute_error: 0.2407\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0085 - mean_absolute_error: 0.0630 - val_loss: 0.2249 - val_mean_absolute_error: 0.2619\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0080 - mean_absolute_error: 0.0643 - val_loss: 0.1999 - val_mean_absolute_error: 0.2297\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0062 - mean_absolute_error: 0.0535 - val_loss: 0.1972 - val_mean_absolute_error: 0.2489\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0061 - mean_absolute_error: 0.0538 - val_loss: 0.2068 - val_mean_absolute_error: 0.2284\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0040 - mean_absolute_error: 0.0428 - val_loss: 0.2116 - val_mean_absolute_error: 0.2358\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0037 - mean_absolute_error: 0.0415 - val_loss: 0.1950 - val_mean_absolute_error: 0.2243\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0029 - mean_absolute_error: 0.0387 - val_loss: 0.1966 - val_mean_absolute_error: 0.2219\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0027 - mean_absolute_error: 0.0350 - val_loss: 0.2084 - val_mean_absolute_error: 0.2268\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0023 - mean_absolute_error: 0.0335 - val_loss: 0.2051 - val_mean_absolute_error: 0.2208\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0022 - mean_absolute_error: 0.0318 - val_loss: 0.1949 - val_mean_absolute_error: 0.2219\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0022 - mean_absolute_error: 0.0318 - val_loss: 0.1977 - val_mean_absolute_error: 0.2165\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0021 - mean_absolute_error: 0.0306 - val_loss: 0.2083 - val_mean_absolute_error: 0.2245\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0020 - mean_absolute_error: 0.0310 - val_loss: 0.2058 - val_mean_absolute_error: 0.2231\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0023 - mean_absolute_error: 0.0328 - val_loss: 0.1945 - val_mean_absolute_error: 0.2213\n",
      "Epoch 47/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0023 - mean_absolute_error: 0.0338 - val_loss: 0.1977 - val_mean_absolute_error: 0.2245\n",
      "Epoch 48/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0028 - mean_absolute_error: 0.0365 - val_loss: 0.2077 - val_mean_absolute_error: 0.2232\n",
      "Epoch 49/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0023 - mean_absolute_error: 0.0335 - val_loss: 0.2111 - val_mean_absolute_error: 0.2352\n",
      "Epoch 50/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0034 - mean_absolute_error: 0.0412 - val_loss: 0.1937 - val_mean_absolute_error: 0.2153\n",
      "Epoch 51/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0030 - mean_absolute_error: 0.0387 - val_loss: 0.1979 - val_mean_absolute_error: 0.2375\n",
      "Epoch 52/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0045 - mean_absolute_error: 0.0477 - val_loss: 0.2064 - val_mean_absolute_error: 0.2200\n",
      "Epoch 53/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0031 - mean_absolute_error: 0.0390 - val_loss: 0.2177 - val_mean_absolute_error: 0.2467\n",
      "Epoch 54/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0049 - mean_absolute_error: 0.0500 - val_loss: 0.1932 - val_mean_absolute_error: 0.2100\n",
      "Epoch 55/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0040 - mean_absolute_error: 0.0439 - val_loss: 0.1973 - val_mean_absolute_error: 0.2448\n",
      "Epoch 56/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0059 - mean_absolute_error: 0.0561 - val_loss: 0.2059 - val_mean_absolute_error: 0.2198\n",
      "Epoch 57/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0038 - mean_absolute_error: 0.0442 - val_loss: 0.2207 - val_mean_absolute_error: 0.2496\n",
      "Epoch 58/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0055 - mean_absolute_error: 0.0539 - val_loss: 0.1917 - val_mean_absolute_error: 0.2113\n",
      "Epoch 59/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0042 - mean_absolute_error: 0.0460 - val_loss: 0.1957 - val_mean_absolute_error: 0.2423\n",
      "Epoch 60/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0056 - mean_absolute_error: 0.0552 - val_loss: 0.2086 - val_mean_absolute_error: 0.2268\n",
      "Epoch 61/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0036 - mean_absolute_error: 0.0441 - val_loss: 0.2173 - val_mean_absolute_error: 0.2430\n",
      "Epoch 62/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0046 - mean_absolute_error: 0.0506 - val_loss: 0.1908 - val_mean_absolute_error: 0.2195\n",
      "Epoch 63/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0037 - mean_absolute_error: 0.0453 - val_loss: 0.1942 - val_mean_absolute_error: 0.2291\n",
      "Epoch 64/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0039 - mean_absolute_error: 0.0458 - val_loss: 0.2132 - val_mean_absolute_error: 0.2352\n",
      "Epoch 65/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0032 - mean_absolute_error: 0.0430 - val_loss: 0.2082 - val_mean_absolute_error: 0.2238\n",
      "Epoch 66/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0031 - mean_absolute_error: 0.0413 - val_loss: 0.1922 - val_mean_absolute_error: 0.2275\n",
      "Epoch 67/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0033 - mean_absolute_error: 0.0442 - val_loss: 0.1959 - val_mean_absolute_error: 0.2127\n",
      "Epoch 68/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0023 - mean_absolute_error: 0.0363 - val_loss: 0.2147 - val_mean_absolute_error: 0.2362\n",
      "Epoch 69/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0027 - mean_absolute_error: 0.0403 - val_loss: 0.2005 - val_mean_absolute_error: 0.2048\n",
      "Epoch 70/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0020 - mean_absolute_error: 0.0316 - val_loss: 0.1938 - val_mean_absolute_error: 0.2264\n",
      "Epoch 71/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0026 - mean_absolute_error: 0.0390 - val_loss: 0.2002 - val_mean_absolute_error: 0.2069\n",
      "Epoch 72/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0016 - mean_absolute_error: 0.0294 - val_loss: 0.2119 - val_mean_absolute_error: 0.2303\n",
      "2/2 [==============================] - 0s 66ms/step - loss: 0.2119 - mean_absolute_error: 0.2303\n",
      "2308\n",
      "Done with seed: 18\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 307ms/step - loss: 1.0687 - mean_absolute_error: 0.8097 - val_loss: 0.3923 - val_mean_absolute_error: 0.5326\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 231ms/step - loss: 0.4229 - mean_absolute_error: 0.5151 - val_loss: 0.3152 - val_mean_absolute_error: 0.4590\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 227ms/step - loss: 0.2998 - mean_absolute_error: 0.4164 - val_loss: 0.2275 - val_mean_absolute_error: 0.3779\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 2s 227ms/step - loss: 0.2262 - mean_absolute_error: 0.3559 - val_loss: 0.2051 - val_mean_absolute_error: 0.3423\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 210ms/step - loss: 0.1809 - mean_absolute_error: 0.3020 - val_loss: 0.1764 - val_mean_absolute_error: 0.3062\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 218ms/step - loss: 0.1393 - mean_absolute_error: 0.2642 - val_loss: 0.1566 - val_mean_absolute_error: 0.2894\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 2s 246ms/step - loss: 0.1265 - mean_absolute_error: 0.2467 - val_loss: 0.1541 - val_mean_absolute_error: 0.2631\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 246ms/step - loss: 0.1135 - mean_absolute_error: 0.2410 - val_loss: 0.1581 - val_mean_absolute_error: 0.2622\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0954 - mean_absolute_error: 0.2121 - val_loss: 0.1365 - val_mean_absolute_error: 0.2378\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 199ms/step - loss: 0.0667 - mean_absolute_error: 0.1759 - val_loss: 0.1266 - val_mean_absolute_error: 0.2272\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.0539 - mean_absolute_error: 0.1575 - val_loss: 0.1094 - val_mean_absolute_error: 0.2132\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0433 - mean_absolute_error: 0.1413 - val_loss: 0.1230 - val_mean_absolute_error: 0.2167\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0396 - mean_absolute_error: 0.1377 - val_loss: 0.1523 - val_mean_absolute_error: 0.2517\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0384 - mean_absolute_error: 0.1393 - val_loss: 0.1203 - val_mean_absolute_error: 0.2062\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0358 - mean_absolute_error: 0.1263 - val_loss: 0.1272 - val_mean_absolute_error: 0.2300\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0356 - mean_absolute_error: 0.1323 - val_loss: 0.1007 - val_mean_absolute_error: 0.1983\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0308 - mean_absolute_error: 0.1242 - val_loss: 0.1128 - val_mean_absolute_error: 0.1992\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0301 - mean_absolute_error: 0.1165 - val_loss: 0.0974 - val_mean_absolute_error: 0.1878\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0253 - mean_absolute_error: 0.1128 - val_loss: 0.1063 - val_mean_absolute_error: 0.1970\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0156 - mean_absolute_error: 0.0844 - val_loss: 0.1094 - val_mean_absolute_error: 0.1930\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0149 - mean_absolute_error: 0.0818 - val_loss: 0.1146 - val_mean_absolute_error: 0.1832\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0155 - mean_absolute_error: 0.0797 - val_loss: 0.1195 - val_mean_absolute_error: 0.1918\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0121 - mean_absolute_error: 0.0737 - val_loss: 0.1169 - val_mean_absolute_error: 0.1759\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0112 - mean_absolute_error: 0.0717 - val_loss: 0.1051 - val_mean_absolute_error: 0.1689\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0101 - mean_absolute_error: 0.0664 - val_loss: 0.1104 - val_mean_absolute_error: 0.1675\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0070 - mean_absolute_error: 0.0534 - val_loss: 0.1220 - val_mean_absolute_error: 0.1872\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0065 - mean_absolute_error: 0.0532 - val_loss: 0.1122 - val_mean_absolute_error: 0.1679\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0048 - mean_absolute_error: 0.0476 - val_loss: 0.1060 - val_mean_absolute_error: 0.1626\n",
      "2/2 [==============================] - 0s 70ms/step - loss: 0.1060 - mean_absolute_error: 0.1626\n",
      "2308\n",
      "Done with seed: 19\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 235ms/step - loss: 0.9639 - mean_absolute_error: 0.7941 - val_loss: 0.5337 - val_mean_absolute_error: 0.5809\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.4843 - mean_absolute_error: 0.5369 - val_loss: 0.5061 - val_mean_absolute_error: 0.5482\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 196ms/step - loss: 0.3571 - mean_absolute_error: 0.4731 - val_loss: 0.4700 - val_mean_absolute_error: 0.5046\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.2497 - mean_absolute_error: 0.3814 - val_loss: 0.3986 - val_mean_absolute_error: 0.4490\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.1904 - mean_absolute_error: 0.3254 - val_loss: 0.4312 - val_mean_absolute_error: 0.4525\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 183ms/step - loss: 0.1629 - mean_absolute_error: 0.3027 - val_loss: 0.3732 - val_mean_absolute_error: 0.4349\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 183ms/step - loss: 0.1336 - mean_absolute_error: 0.2689 - val_loss: 0.2906 - val_mean_absolute_error: 0.3825\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.1030 - mean_absolute_error: 0.2337 - val_loss: 0.2545 - val_mean_absolute_error: 0.3354\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.0821 - mean_absolute_error: 0.2020 - val_loss: 0.2629 - val_mean_absolute_error: 0.3227\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0679 - mean_absolute_error: 0.1798 - val_loss: 0.2273 - val_mean_absolute_error: 0.2944\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0535 - mean_absolute_error: 0.1580 - val_loss: 0.2084 - val_mean_absolute_error: 0.2627\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.0361 - mean_absolute_error: 0.1285 - val_loss: 0.2235 - val_mean_absolute_error: 0.2652\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.0319 - mean_absolute_error: 0.1237 - val_loss: 0.1998 - val_mean_absolute_error: 0.2581\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0353 - mean_absolute_error: 0.1222 - val_loss: 0.1870 - val_mean_absolute_error: 0.2857\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0441 - mean_absolute_error: 0.1480 - val_loss: 0.2240 - val_mean_absolute_error: 0.2612\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0537 - mean_absolute_error: 0.1509 - val_loss: 0.2660 - val_mean_absolute_error: 0.3291\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.0450 - mean_absolute_error: 0.1498 - val_loss: 0.1966 - val_mean_absolute_error: 0.2648\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0330 - mean_absolute_error: 0.1245 - val_loss: 0.2131 - val_mean_absolute_error: 0.2688\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.0279 - mean_absolute_error: 0.1215 - val_loss: 0.1931 - val_mean_absolute_error: 0.2452\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0207 - mean_absolute_error: 0.0978 - val_loss: 0.2084 - val_mean_absolute_error: 0.2628\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0145 - mean_absolute_error: 0.0913 - val_loss: 0.2075 - val_mean_absolute_error: 0.2356\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0125 - mean_absolute_error: 0.0778 - val_loss: 0.1884 - val_mean_absolute_error: 0.2259\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0138 - mean_absolute_error: 0.0814 - val_loss: 0.1878 - val_mean_absolute_error: 0.2198\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0127 - mean_absolute_error: 0.0767 - val_loss: 0.2017 - val_mean_absolute_error: 0.2370\n",
      "2/2 [==============================] - 0s 79ms/step - loss: 0.2017 - mean_absolute_error: 0.2370\n",
      "2308\n",
      "Done with seed: 20\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 352ms/step - loss: 1.3940 - mean_absolute_error: 0.9354 - val_loss: 0.7316 - val_mean_absolute_error: 0.6761\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 272ms/step - loss: 0.5141 - mean_absolute_error: 0.5498 - val_loss: 0.4971 - val_mean_absolute_error: 0.5362\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 285ms/step - loss: 0.3665 - mean_absolute_error: 0.4496 - val_loss: 0.3763 - val_mean_absolute_error: 0.4721\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 2s 363ms/step - loss: 0.3090 - mean_absolute_error: 0.4034 - val_loss: 0.3247 - val_mean_absolute_error: 0.4424\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 2s 248ms/step - loss: 0.2781 - mean_absolute_error: 0.3822 - val_loss: 0.2875 - val_mean_absolute_error: 0.4061\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 229ms/step - loss: 0.2389 - mean_absolute_error: 0.3446 - val_loss: 0.2770 - val_mean_absolute_error: 0.3866\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 220ms/step - loss: 0.1976 - mean_absolute_error: 0.3085 - val_loss: 0.2480 - val_mean_absolute_error: 0.3536\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 197ms/step - loss: 0.1656 - mean_absolute_error: 0.2779 - val_loss: 0.2274 - val_mean_absolute_error: 0.3263\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 188ms/step - loss: 0.1471 - mean_absolute_error: 0.2536 - val_loss: 0.1890 - val_mean_absolute_error: 0.2983\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.1272 - mean_absolute_error: 0.2318 - val_loss: 0.1827 - val_mean_absolute_error: 0.2874\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.1061 - mean_absolute_error: 0.2098 - val_loss: 0.1623 - val_mean_absolute_error: 0.2683\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0892 - mean_absolute_error: 0.1878 - val_loss: 0.1741 - val_mean_absolute_error: 0.2697\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0749 - mean_absolute_error: 0.1663 - val_loss: 0.1598 - val_mean_absolute_error: 0.2562\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0637 - mean_absolute_error: 0.1485 - val_loss: 0.1660 - val_mean_absolute_error: 0.2538\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0581 - mean_absolute_error: 0.1409 - val_loss: 0.1278 - val_mean_absolute_error: 0.2296\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0542 - mean_absolute_error: 0.1346 - val_loss: 0.1541 - val_mean_absolute_error: 0.2385\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0483 - mean_absolute_error: 0.1245 - val_loss: 0.1558 - val_mean_absolute_error: 0.2335\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0419 - mean_absolute_error: 0.1136 - val_loss: 0.1630 - val_mean_absolute_error: 0.2312\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0377 - mean_absolute_error: 0.1068 - val_loss: 0.1612 - val_mean_absolute_error: 0.2337\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0374 - mean_absolute_error: 0.1095 - val_loss: 0.1567 - val_mean_absolute_error: 0.2201\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0331 - mean_absolute_error: 0.1041 - val_loss: 0.1529 - val_mean_absolute_error: 0.2351\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0298 - mean_absolute_error: 0.0999 - val_loss: 0.1606 - val_mean_absolute_error: 0.2257\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0274 - mean_absolute_error: 0.0958 - val_loss: 0.1700 - val_mean_absolute_error: 0.2480\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0293 - mean_absolute_error: 0.1100 - val_loss: 0.1439 - val_mean_absolute_error: 0.2367\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0342 - mean_absolute_error: 0.1166 - val_loss: 0.1826 - val_mean_absolute_error: 0.2940\n",
      "2/2 [==============================] - 0s 66ms/step - loss: 0.1826 - mean_absolute_error: 0.2940\n",
      "2308\n",
      "Done with seed: 21\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 221ms/step - loss: 0.7798 - mean_absolute_error: 0.6800 - val_loss: 0.7624 - val_mean_absolute_error: 0.7342\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.5016 - mean_absolute_error: 0.5755 - val_loss: 0.5601 - val_mean_absolute_error: 0.5462\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.3519 - mean_absolute_error: 0.4457 - val_loss: 0.5036 - val_mean_absolute_error: 0.4893\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.2734 - mean_absolute_error: 0.3910 - val_loss: 0.4237 - val_mean_absolute_error: 0.4606\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 213ms/step - loss: 0.2457 - mean_absolute_error: 0.3655 - val_loss: 0.3926 - val_mean_absolute_error: 0.4110\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 213ms/step - loss: 0.2025 - mean_absolute_error: 0.3256 - val_loss: 0.3447 - val_mean_absolute_error: 0.3900\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 207ms/step - loss: 0.1809 - mean_absolute_error: 0.3060 - val_loss: 0.3223 - val_mean_absolute_error: 0.3749\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 203ms/step - loss: 0.1613 - mean_absolute_error: 0.2858 - val_loss: 0.2972 - val_mean_absolute_error: 0.3536\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 212ms/step - loss: 0.1479 - mean_absolute_error: 0.2627 - val_loss: 0.2527 - val_mean_absolute_error: 0.3463\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 2s 213ms/step - loss: 0.1276 - mean_absolute_error: 0.2461 - val_loss: 0.2420 - val_mean_absolute_error: 0.3249\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 217ms/step - loss: 0.1039 - mean_absolute_error: 0.2281 - val_loss: 0.2008 - val_mean_absolute_error: 0.2935\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 211ms/step - loss: 0.0953 - mean_absolute_error: 0.2100 - val_loss: 0.1938 - val_mean_absolute_error: 0.2763\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 2s 225ms/step - loss: 0.0723 - mean_absolute_error: 0.1833 - val_loss: 0.1787 - val_mean_absolute_error: 0.2604\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 212ms/step - loss: 0.0630 - mean_absolute_error: 0.1651 - val_loss: 0.1762 - val_mean_absolute_error: 0.2520\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0518 - mean_absolute_error: 0.1507 - val_loss: 0.1686 - val_mean_absolute_error: 0.2413\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.0461 - mean_absolute_error: 0.1428 - val_loss: 0.1575 - val_mean_absolute_error: 0.2391\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0577 - mean_absolute_error: 0.1515 - val_loss: 0.1476 - val_mean_absolute_error: 0.2409\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0443 - mean_absolute_error: 0.1414 - val_loss: 0.1398 - val_mean_absolute_error: 0.2268\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0329 - mean_absolute_error: 0.1237 - val_loss: 0.1228 - val_mean_absolute_error: 0.2054\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0231 - mean_absolute_error: 0.1077 - val_loss: 0.1131 - val_mean_absolute_error: 0.1960\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0185 - mean_absolute_error: 0.0884 - val_loss: 0.1164 - val_mean_absolute_error: 0.1927\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0145 - mean_absolute_error: 0.0768 - val_loss: 0.1233 - val_mean_absolute_error: 0.1831\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0108 - mean_absolute_error: 0.0691 - val_loss: 0.1120 - val_mean_absolute_error: 0.1722\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0102 - mean_absolute_error: 0.0642 - val_loss: 0.1073 - val_mean_absolute_error: 0.1704\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 197ms/step - loss: 0.0084 - mean_absolute_error: 0.0550 - val_loss: 0.1136 - val_mean_absolute_error: 0.1697\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 216ms/step - loss: 0.0063 - mean_absolute_error: 0.0460 - val_loss: 0.1159 - val_mean_absolute_error: 0.1594\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 2s 251ms/step - loss: 0.0048 - mean_absolute_error: 0.0423 - val_loss: 0.1122 - val_mean_absolute_error: 0.1531\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 2s 308ms/step - loss: 0.0051 - mean_absolute_error: 0.0444 - val_loss: 0.1044 - val_mean_absolute_error: 0.1537\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 2s 346ms/step - loss: 0.0065 - mean_absolute_error: 0.0497 - val_loss: 0.1097 - val_mean_absolute_error: 0.1778\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 2s 333ms/step - loss: 0.0070 - mean_absolute_error: 0.0577 - val_loss: 0.1176 - val_mean_absolute_error: 0.1656\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 2s 351ms/step - loss: 0.0067 - mean_absolute_error: 0.0563 - val_loss: 0.1271 - val_mean_absolute_error: 0.1853\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 2s 264ms/step - loss: 0.0074 - mean_absolute_error: 0.0666 - val_loss: 0.1082 - val_mean_absolute_error: 0.1570\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 2s 289ms/step - loss: 0.0083 - mean_absolute_error: 0.0652 - val_loss: 0.1013 - val_mean_absolute_error: 0.1902\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 2s 293ms/step - loss: 0.0118 - mean_absolute_error: 0.0779 - val_loss: 0.1039 - val_mean_absolute_error: 0.1733\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 204ms/step - loss: 0.0097 - mean_absolute_error: 0.0716 - val_loss: 0.1353 - val_mean_absolute_error: 0.1984\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 216ms/step - loss: 0.0112 - mean_absolute_error: 0.0792 - val_loss: 0.1104 - val_mean_absolute_error: 0.1637\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.0108 - mean_absolute_error: 0.0766 - val_loss: 0.1023 - val_mean_absolute_error: 0.1996\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0118 - mean_absolute_error: 0.0821 - val_loss: 0.1403 - val_mean_absolute_error: 0.1718\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0054 - mean_absolute_error: 0.0512 - val_loss: 0.1365 - val_mean_absolute_error: 0.1965\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0159 - mean_absolute_error: 0.0820 - val_loss: 0.0893 - val_mean_absolute_error: 0.1749\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0160 - mean_absolute_error: 0.0886 - val_loss: 0.0963 - val_mean_absolute_error: 0.1849\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0138 - mean_absolute_error: 0.0876 - val_loss: 0.1623 - val_mean_absolute_error: 0.2187\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0138 - mean_absolute_error: 0.0891 - val_loss: 0.1290 - val_mean_absolute_error: 0.1846\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0172 - mean_absolute_error: 0.0885 - val_loss: 0.0990 - val_mean_absolute_error: 0.1887\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0134 - mean_absolute_error: 0.0880 - val_loss: 0.1197 - val_mean_absolute_error: 0.1840\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0091 - mean_absolute_error: 0.0672 - val_loss: 0.1133 - val_mean_absolute_error: 0.1712\n",
      "Epoch 47/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0089 - mean_absolute_error: 0.0625 - val_loss: 0.1007 - val_mean_absolute_error: 0.1646\n",
      "Epoch 48/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0069 - mean_absolute_error: 0.0561 - val_loss: 0.1202 - val_mean_absolute_error: 0.1616\n",
      "Epoch 49/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0049 - mean_absolute_error: 0.0438 - val_loss: 0.1207 - val_mean_absolute_error: 0.1587\n",
      "Epoch 50/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0030 - mean_absolute_error: 0.0374 - val_loss: 0.1100 - val_mean_absolute_error: 0.1522\n",
      "2/2 [==============================] - 0s 61ms/step - loss: 0.1100 - mean_absolute_error: 0.1522\n",
      "2308\n",
      "Done with seed: 22\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 198ms/step - loss: 0.9661 - mean_absolute_error: 0.7857 - val_loss: 0.4495 - val_mean_absolute_error: 0.5120\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.5358 - mean_absolute_error: 0.5581 - val_loss: 0.3486 - val_mean_absolute_error: 0.4805\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.2945 - mean_absolute_error: 0.4106 - val_loss: 0.2569 - val_mean_absolute_error: 0.4000\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.2435 - mean_absolute_error: 0.3787 - val_loss: 0.2479 - val_mean_absolute_error: 0.3815\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.1783 - mean_absolute_error: 0.3090 - val_loss: 0.2171 - val_mean_absolute_error: 0.3391\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.1291 - mean_absolute_error: 0.2650 - val_loss: 0.1958 - val_mean_absolute_error: 0.3118\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0928 - mean_absolute_error: 0.2078 - val_loss: 0.2038 - val_mean_absolute_error: 0.3111\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0703 - mean_absolute_error: 0.1803 - val_loss: 0.1747 - val_mean_absolute_error: 0.2805\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 135ms/step - loss: 0.0471 - mean_absolute_error: 0.1431 - val_loss: 0.1605 - val_mean_absolute_error: 0.2669\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0390 - mean_absolute_error: 0.1227 - val_loss: 0.1535 - val_mean_absolute_error: 0.2593\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0321 - mean_absolute_error: 0.1120 - val_loss: 0.1440 - val_mean_absolute_error: 0.2398\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0241 - mean_absolute_error: 0.1059 - val_loss: 0.1486 - val_mean_absolute_error: 0.2434\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0273 - mean_absolute_error: 0.1021 - val_loss: 0.1364 - val_mean_absolute_error: 0.2343\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0202 - mean_absolute_error: 0.0858 - val_loss: 0.1363 - val_mean_absolute_error: 0.2289\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0135 - mean_absolute_error: 0.0670 - val_loss: 0.1337 - val_mean_absolute_error: 0.2202\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0124 - mean_absolute_error: 0.0721 - val_loss: 0.1317 - val_mean_absolute_error: 0.2109\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0113 - mean_absolute_error: 0.0666 - val_loss: 0.1370 - val_mean_absolute_error: 0.2330\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0129 - mean_absolute_error: 0.0734 - val_loss: 0.1387 - val_mean_absolute_error: 0.2352\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0174 - mean_absolute_error: 0.0882 - val_loss: 0.1386 - val_mean_absolute_error: 0.2161\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0155 - mean_absolute_error: 0.0859 - val_loss: 0.1507 - val_mean_absolute_error: 0.2420\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0190 - mean_absolute_error: 0.1003 - val_loss: 0.1307 - val_mean_absolute_error: 0.2151\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0136 - mean_absolute_error: 0.0788 - val_loss: 0.1359 - val_mean_absolute_error: 0.2351\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0195 - mean_absolute_error: 0.0997 - val_loss: 0.1465 - val_mean_absolute_error: 0.2284\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0118 - mean_absolute_error: 0.0753 - val_loss: 0.1372 - val_mean_absolute_error: 0.2083\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0112 - mean_absolute_error: 0.0696 - val_loss: 0.1289 - val_mean_absolute_error: 0.2202\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0100 - mean_absolute_error: 0.0656 - val_loss: 0.1341 - val_mean_absolute_error: 0.2035\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0088 - mean_absolute_error: 0.0568 - val_loss: 0.1418 - val_mean_absolute_error: 0.2144\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0096 - mean_absolute_error: 0.0666 - val_loss: 0.1310 - val_mean_absolute_error: 0.2073\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0069 - mean_absolute_error: 0.0508 - val_loss: 0.1291 - val_mean_absolute_error: 0.2089\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 226ms/step - loss: 0.0078 - mean_absolute_error: 0.0558 - val_loss: 0.1352 - val_mean_absolute_error: 0.2016\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 2s 235ms/step - loss: 0.0054 - mean_absolute_error: 0.0413 - val_loss: 0.1325 - val_mean_absolute_error: 0.1937\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0054 - mean_absolute_error: 0.0421 - val_loss: 0.1256 - val_mean_absolute_error: 0.1972\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.0046 - mean_absolute_error: 0.0373 - val_loss: 0.1273 - val_mean_absolute_error: 0.1900\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0047 - mean_absolute_error: 0.0364 - val_loss: 0.1348 - val_mean_absolute_error: 0.1999\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0041 - mean_absolute_error: 0.0331 - val_loss: 0.1287 - val_mean_absolute_error: 0.1856\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0039 - mean_absolute_error: 0.0305 - val_loss: 0.1251 - val_mean_absolute_error: 0.1946\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0040 - mean_absolute_error: 0.0331 - val_loss: 0.1289 - val_mean_absolute_error: 0.1890\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0040 - mean_absolute_error: 0.0326 - val_loss: 0.1340 - val_mean_absolute_error: 0.1973\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0038 - mean_absolute_error: 0.0323 - val_loss: 0.1294 - val_mean_absolute_error: 0.1857\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0038 - mean_absolute_error: 0.0324 - val_loss: 0.1261 - val_mean_absolute_error: 0.1958\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0039 - mean_absolute_error: 0.0347 - val_loss: 0.1282 - val_mean_absolute_error: 0.1957\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0046 - mean_absolute_error: 0.0408 - val_loss: 0.1333 - val_mean_absolute_error: 0.1946\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0039 - mean_absolute_error: 0.0352 - val_loss: 0.1328 - val_mean_absolute_error: 0.1973\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0051 - mean_absolute_error: 0.0461 - val_loss: 0.1256 - val_mean_absolute_error: 0.1903\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0046 - mean_absolute_error: 0.0428 - val_loss: 0.1275 - val_mean_absolute_error: 0.2103\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0067 - mean_absolute_error: 0.0570 - val_loss: 0.1320 - val_mean_absolute_error: 0.1904\n",
      "2/2 [==============================] - 0s 76ms/step - loss: 0.1320 - mean_absolute_error: 0.1904\n",
      "2308\n",
      "Done with seed: 23\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 354ms/step - loss: 0.7280 - mean_absolute_error: 0.6779 - val_loss: 0.4531 - val_mean_absolute_error: 0.5303\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.3643 - mean_absolute_error: 0.4830 - val_loss: 0.3328 - val_mean_absolute_error: 0.4632\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.2421 - mean_absolute_error: 0.3925 - val_loss: 0.1823 - val_mean_absolute_error: 0.3311\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.1520 - mean_absolute_error: 0.2949 - val_loss: 0.2010 - val_mean_absolute_error: 0.3519\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.1055 - mean_absolute_error: 0.2410 - val_loss: 0.2105 - val_mean_absolute_error: 0.3233\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0787 - mean_absolute_error: 0.2047 - val_loss: 0.1855 - val_mean_absolute_error: 0.2881\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 124ms/step - loss: 0.0626 - mean_absolute_error: 0.1728 - val_loss: 0.1549 - val_mean_absolute_error: 0.2633\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0730 - mean_absolute_error: 0.1819 - val_loss: 0.2204 - val_mean_absolute_error: 0.3439\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 125ms/step - loss: 0.1048 - mean_absolute_error: 0.2398 - val_loss: 0.2349 - val_mean_absolute_error: 0.3200\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0612 - mean_absolute_error: 0.1690 - val_loss: 0.1230 - val_mean_absolute_error: 0.2307\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0453 - mean_absolute_error: 0.1433 - val_loss: 0.1234 - val_mean_absolute_error: 0.2199\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0332 - mean_absolute_error: 0.1276 - val_loss: 0.1383 - val_mean_absolute_error: 0.2314\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0250 - mean_absolute_error: 0.1117 - val_loss: 0.1208 - val_mean_absolute_error: 0.2054\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 135ms/step - loss: 0.0172 - mean_absolute_error: 0.0841 - val_loss: 0.1161 - val_mean_absolute_error: 0.2008\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0132 - mean_absolute_error: 0.0793 - val_loss: 0.1244 - val_mean_absolute_error: 0.1952\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0107 - mean_absolute_error: 0.0654 - val_loss: 0.1457 - val_mean_absolute_error: 0.2007\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0106 - mean_absolute_error: 0.0658 - val_loss: 0.1379 - val_mean_absolute_error: 0.1888\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0073 - mean_absolute_error: 0.0557 - val_loss: 0.1353 - val_mean_absolute_error: 0.1927\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0068 - mean_absolute_error: 0.0559 - val_loss: 0.1396 - val_mean_absolute_error: 0.1835\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0053 - mean_absolute_error: 0.0485 - val_loss: 0.1519 - val_mean_absolute_error: 0.2062\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0065 - mean_absolute_error: 0.0584 - val_loss: 0.1497 - val_mean_absolute_error: 0.1895\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0058 - mean_absolute_error: 0.0497 - val_loss: 0.1396 - val_mean_absolute_error: 0.1869\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0051 - mean_absolute_error: 0.0509 - val_loss: 0.1344 - val_mean_absolute_error: 0.1995\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0074 - mean_absolute_error: 0.0620 - val_loss: 0.1369 - val_mean_absolute_error: 0.1765\n",
      "2/2 [==============================] - 0s 71ms/step - loss: 0.1369 - mean_absolute_error: 0.1765\n",
      "2308\n",
      "Done with seed: 24\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 241ms/step - loss: 0.8766 - mean_absolute_error: 0.7435 - val_loss: 0.6596 - val_mean_absolute_error: 0.6379\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.4518 - mean_absolute_error: 0.5156 - val_loss: 0.5536 - val_mean_absolute_error: 0.5991\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 222ms/step - loss: 0.3168 - mean_absolute_error: 0.4358 - val_loss: 0.4668 - val_mean_absolute_error: 0.5067\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 220ms/step - loss: 0.2486 - mean_absolute_error: 0.3739 - val_loss: 0.4036 - val_mean_absolute_error: 0.4847\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 219ms/step - loss: 0.1736 - mean_absolute_error: 0.3090 - val_loss: 0.3309 - val_mean_absolute_error: 0.4215\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 222ms/step - loss: 0.1337 - mean_absolute_error: 0.2681 - val_loss: 0.3004 - val_mean_absolute_error: 0.3835\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.1047 - mean_absolute_error: 0.2306 - val_loss: 0.2506 - val_mean_absolute_error: 0.3522\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 306ms/step - loss: 0.0751 - mean_absolute_error: 0.1940 - val_loss: 0.2469 - val_mean_absolute_error: 0.3169\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 2s 253ms/step - loss: 0.0606 - mean_absolute_error: 0.1759 - val_loss: 0.2377 - val_mean_absolute_error: 0.3059\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 2s 262ms/step - loss: 0.0516 - mean_absolute_error: 0.1660 - val_loss: 0.1970 - val_mean_absolute_error: 0.2792\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 307ms/step - loss: 0.0345 - mean_absolute_error: 0.1321 - val_loss: 0.1927 - val_mean_absolute_error: 0.2697\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 2s 346ms/step - loss: 0.0248 - mean_absolute_error: 0.1097 - val_loss: 0.1808 - val_mean_absolute_error: 0.2517\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 2s 365ms/step - loss: 0.0256 - mean_absolute_error: 0.1097 - val_loss: 0.1733 - val_mean_absolute_error: 0.2457\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 2s 321ms/step - loss: 0.0242 - mean_absolute_error: 0.1089 - val_loss: 0.1914 - val_mean_absolute_error: 0.2666\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 219ms/step - loss: 0.0255 - mean_absolute_error: 0.1131 - val_loss: 0.1983 - val_mean_absolute_error: 0.2328\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 227ms/step - loss: 0.0222 - mean_absolute_error: 0.1004 - val_loss: 0.2204 - val_mean_absolute_error: 0.2576\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0169 - mean_absolute_error: 0.0940 - val_loss: 0.2375 - val_mean_absolute_error: 0.2586\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 194ms/step - loss: 0.0140 - mean_absolute_error: 0.0812 - val_loss: 0.1838 - val_mean_absolute_error: 0.2245\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0128 - mean_absolute_error: 0.0813 - val_loss: 0.1845 - val_mean_absolute_error: 0.2425\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0118 - mean_absolute_error: 0.0759 - val_loss: 0.1773 - val_mean_absolute_error: 0.2229\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0141 - mean_absolute_error: 0.0842 - val_loss: 0.2049 - val_mean_absolute_error: 0.2318\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0138 - mean_absolute_error: 0.0818 - val_loss: 0.2486 - val_mean_absolute_error: 0.2726\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0180 - mean_absolute_error: 0.0941 - val_loss: 0.1838 - val_mean_absolute_error: 0.2265\n",
      "2/2 [==============================] - 0s 64ms/step - loss: 0.1838 - mean_absolute_error: 0.2265\n",
      "2308\n",
      "Done with seed: 25\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 220ms/step - loss: 0.8595 - mean_absolute_error: 0.7292 - val_loss: 0.5805 - val_mean_absolute_error: 0.5749\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.4495 - mean_absolute_error: 0.5194 - val_loss: 0.4383 - val_mean_absolute_error: 0.4952\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.3093 - mean_absolute_error: 0.4180 - val_loss: 0.4227 - val_mean_absolute_error: 0.4851\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.2185 - mean_absolute_error: 0.3390 - val_loss: 0.3667 - val_mean_absolute_error: 0.4416\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.1888 - mean_absolute_error: 0.3165 - val_loss: 0.3150 - val_mean_absolute_error: 0.3952\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.1441 - mean_absolute_error: 0.2668 - val_loss: 0.2833 - val_mean_absolute_error: 0.3567\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.1022 - mean_absolute_error: 0.2198 - val_loss: 0.2810 - val_mean_absolute_error: 0.3504\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0813 - mean_absolute_error: 0.2004 - val_loss: 0.2489 - val_mean_absolute_error: 0.3206\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0744 - mean_absolute_error: 0.1887 - val_loss: 0.2352 - val_mean_absolute_error: 0.2903\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0539 - mean_absolute_error: 0.1566 - val_loss: 0.2416 - val_mean_absolute_error: 0.2873\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0424 - mean_absolute_error: 0.1416 - val_loss: 0.2382 - val_mean_absolute_error: 0.2741\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0377 - mean_absolute_error: 0.1335 - val_loss: 0.2401 - val_mean_absolute_error: 0.2684\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0365 - mean_absolute_error: 0.1304 - val_loss: 0.2524 - val_mean_absolute_error: 0.2714\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0238 - mean_absolute_error: 0.1039 - val_loss: 0.2357 - val_mean_absolute_error: 0.2583\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0177 - mean_absolute_error: 0.0903 - val_loss: 0.2362 - val_mean_absolute_error: 0.2408\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0133 - mean_absolute_error: 0.0704 - val_loss: 0.2423 - val_mean_absolute_error: 0.2458\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0148 - mean_absolute_error: 0.0780 - val_loss: 0.2462 - val_mean_absolute_error: 0.2397\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0161 - mean_absolute_error: 0.0824 - val_loss: 0.2500 - val_mean_absolute_error: 0.2638\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0170 - mean_absolute_error: 0.0917 - val_loss: 0.2301 - val_mean_absolute_error: 0.2351\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0185 - mean_absolute_error: 0.0941 - val_loss: 0.2590 - val_mean_absolute_error: 0.2780\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0226 - mean_absolute_error: 0.1053 - val_loss: 0.2386 - val_mean_absolute_error: 0.2489\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0223 - mean_absolute_error: 0.1003 - val_loss: 0.2320 - val_mean_absolute_error: 0.2622\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0188 - mean_absolute_error: 0.0999 - val_loss: 0.2290 - val_mean_absolute_error: 0.2489\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0152 - mean_absolute_error: 0.0870 - val_loss: 0.2436 - val_mean_absolute_error: 0.2681\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0282 - mean_absolute_error: 0.1175 - val_loss: 0.2156 - val_mean_absolute_error: 0.2568\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0187 - mean_absolute_error: 0.0975 - val_loss: 0.2252 - val_mean_absolute_error: 0.2599\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 2s 252ms/step - loss: 0.0190 - mean_absolute_error: 0.0993 - val_loss: 0.2559 - val_mean_absolute_error: 0.2839\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 2s 232ms/step - loss: 0.0210 - mean_absolute_error: 0.1059 - val_loss: 0.2100 - val_mean_absolute_error: 0.2224\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 2s 236ms/step - loss: 0.0167 - mean_absolute_error: 0.0881 - val_loss: 0.2435 - val_mean_absolute_error: 0.2565\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 207ms/step - loss: 0.0160 - mean_absolute_error: 0.0905 - val_loss: 0.2482 - val_mean_absolute_error: 0.2618\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 211ms/step - loss: 0.0134 - mean_absolute_error: 0.0824 - val_loss: 0.2311 - val_mean_absolute_error: 0.2224\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0093 - mean_absolute_error: 0.0679 - val_loss: 0.2385 - val_mean_absolute_error: 0.2436\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 220ms/step - loss: 0.0071 - mean_absolute_error: 0.0606 - val_loss: 0.2467 - val_mean_absolute_error: 0.2461\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 2s 250ms/step - loss: 0.0071 - mean_absolute_error: 0.0555 - val_loss: 0.2378 - val_mean_absolute_error: 0.2237\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 2s 235ms/step - loss: 0.0051 - mean_absolute_error: 0.0468 - val_loss: 0.2391 - val_mean_absolute_error: 0.2332\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 2s 249ms/step - loss: 0.0042 - mean_absolute_error: 0.0430 - val_loss: 0.2479 - val_mean_absolute_error: 0.2351\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 2s 259ms/step - loss: 0.0030 - mean_absolute_error: 0.0355 - val_loss: 0.2461 - val_mean_absolute_error: 0.2260\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 2s 255ms/step - loss: 0.0025 - mean_absolute_error: 0.0311 - val_loss: 0.2433 - val_mean_absolute_error: 0.2294\n",
      "2/2 [==============================] - 0s 87ms/step - loss: 0.2433 - mean_absolute_error: 0.2294\n",
      "2308\n",
      "Done with seed: 26\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 5s 285ms/step - loss: 0.9510 - mean_absolute_error: 0.7677 - val_loss: 0.9980 - val_mean_absolute_error: 0.7509\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.6081 - mean_absolute_error: 0.5897 - val_loss: 0.6046 - val_mean_absolute_error: 0.6286\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 182ms/step - loss: 0.3379 - mean_absolute_error: 0.4490 - val_loss: 0.5371 - val_mean_absolute_error: 0.5410\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.2606 - mean_absolute_error: 0.3789 - val_loss: 0.4674 - val_mean_absolute_error: 0.5189\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 183ms/step - loss: 0.2472 - mean_absolute_error: 0.3526 - val_loss: 0.4868 - val_mean_absolute_error: 0.5189\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 192ms/step - loss: 0.1848 - mean_absolute_error: 0.3054 - val_loss: 0.4306 - val_mean_absolute_error: 0.4753\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.1447 - mean_absolute_error: 0.2751 - val_loss: 0.3850 - val_mean_absolute_error: 0.4387\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.1197 - mean_absolute_error: 0.2416 - val_loss: 0.3636 - val_mean_absolute_error: 0.4200\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.0929 - mean_absolute_error: 0.2068 - val_loss: 0.3590 - val_mean_absolute_error: 0.3990\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0937 - mean_absolute_error: 0.2061 - val_loss: 0.3319 - val_mean_absolute_error: 0.3855\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0704 - mean_absolute_error: 0.1730 - val_loss: 0.2610 - val_mean_absolute_error: 0.3217\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0629 - mean_absolute_error: 0.1656 - val_loss: 0.2742 - val_mean_absolute_error: 0.3446\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0532 - mean_absolute_error: 0.1548 - val_loss: 0.2739 - val_mean_absolute_error: 0.3302\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0463 - mean_absolute_error: 0.1411 - val_loss: 0.2650 - val_mean_absolute_error: 0.3153\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0375 - mean_absolute_error: 0.1260 - val_loss: 0.2419 - val_mean_absolute_error: 0.2964\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0330 - mean_absolute_error: 0.1149 - val_loss: 0.2238 - val_mean_absolute_error: 0.2959\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0326 - mean_absolute_error: 0.1101 - val_loss: 0.2238 - val_mean_absolute_error: 0.2850\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0296 - mean_absolute_error: 0.1108 - val_loss: 0.2492 - val_mean_absolute_error: 0.2965\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0234 - mean_absolute_error: 0.0926 - val_loss: 0.2431 - val_mean_absolute_error: 0.2871\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0324 - mean_absolute_error: 0.1151 - val_loss: 0.2178 - val_mean_absolute_error: 0.2860\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0226 - mean_absolute_error: 0.0964 - val_loss: 0.2169 - val_mean_absolute_error: 0.2755\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0254 - mean_absolute_error: 0.1033 - val_loss: 0.2366 - val_mean_absolute_error: 0.2899\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0179 - mean_absolute_error: 0.0840 - val_loss: 0.2157 - val_mean_absolute_error: 0.2628\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0184 - mean_absolute_error: 0.0864 - val_loss: 0.2035 - val_mean_absolute_error: 0.2727\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0180 - mean_absolute_error: 0.0904 - val_loss: 0.2158 - val_mean_absolute_error: 0.2653\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0152 - mean_absolute_error: 0.0779 - val_loss: 0.2325 - val_mean_absolute_error: 0.2928\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0174 - mean_absolute_error: 0.0902 - val_loss: 0.1954 - val_mean_absolute_error: 0.2609\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0145 - mean_absolute_error: 0.0788 - val_loss: 0.1968 - val_mean_absolute_error: 0.2633\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0194 - mean_absolute_error: 0.0951 - val_loss: 0.2443 - val_mean_absolute_error: 0.3168\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0181 - mean_absolute_error: 0.0898 - val_loss: 0.2004 - val_mean_absolute_error: 0.2430\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0092 - mean_absolute_error: 0.0626 - val_loss: 0.1931 - val_mean_absolute_error: 0.2634\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0153 - mean_absolute_error: 0.0825 - val_loss: 0.2130 - val_mean_absolute_error: 0.2591\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0135 - mean_absolute_error: 0.0698 - val_loss: 0.2155 - val_mean_absolute_error: 0.2624\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0113 - mean_absolute_error: 0.0647 - val_loss: 0.1989 - val_mean_absolute_error: 0.2498\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0109 - mean_absolute_error: 0.0678 - val_loss: 0.2187 - val_mean_absolute_error: 0.2584\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0079 - mean_absolute_error: 0.0564 - val_loss: 0.2180 - val_mean_absolute_error: 0.2549\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0060 - mean_absolute_error: 0.0479 - val_loss: 0.2041 - val_mean_absolute_error: 0.2463\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0056 - mean_absolute_error: 0.0495 - val_loss: 0.2078 - val_mean_absolute_error: 0.2382\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0042 - mean_absolute_error: 0.0422 - val_loss: 0.2212 - val_mean_absolute_error: 0.2621\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0048 - mean_absolute_error: 0.0420 - val_loss: 0.2065 - val_mean_absolute_error: 0.2331\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0029 - mean_absolute_error: 0.0297 - val_loss: 0.2019 - val_mean_absolute_error: 0.2389\n",
      "2/2 [==============================] - 0s 65ms/step - loss: 0.2019 - mean_absolute_error: 0.2389\n",
      "2308\n",
      "Done with seed: 27\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 324ms/step - loss: 0.9085 - mean_absolute_error: 0.7834 - val_loss: 0.5637 - val_mean_absolute_error: 0.6193\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 242ms/step - loss: 0.4577 - mean_absolute_error: 0.5428 - val_loss: 0.3970 - val_mean_absolute_error: 0.4918\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 193ms/step - loss: 0.3260 - mean_absolute_error: 0.4248 - val_loss: 0.3654 - val_mean_absolute_error: 0.4669\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 2s 231ms/step - loss: 0.3033 - mean_absolute_error: 0.4013 - val_loss: 0.3396 - val_mean_absolute_error: 0.4452\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 2s 250ms/step - loss: 0.2488 - mean_absolute_error: 0.3680 - val_loss: 0.3033 - val_mean_absolute_error: 0.4044\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 268ms/step - loss: 0.1953 - mean_absolute_error: 0.3211 - val_loss: 0.2713 - val_mean_absolute_error: 0.3710\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 3s 445ms/step - loss: 0.1556 - mean_absolute_error: 0.2824 - val_loss: 0.2339 - val_mean_absolute_error: 0.3530\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 3s 522ms/step - loss: 0.1256 - mean_absolute_error: 0.2509 - val_loss: 0.1979 - val_mean_absolute_error: 0.3150\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 3s 481ms/step - loss: 0.0982 - mean_absolute_error: 0.2253 - val_loss: 0.1688 - val_mean_absolute_error: 0.2736\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 3s 402ms/step - loss: 0.0869 - mean_absolute_error: 0.2137 - val_loss: 0.1721 - val_mean_absolute_error: 0.2745\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 333ms/step - loss: 0.0729 - mean_absolute_error: 0.1950 - val_loss: 0.1461 - val_mean_absolute_error: 0.2428\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 2s 268ms/step - loss: 0.0642 - mean_absolute_error: 0.1766 - val_loss: 0.1419 - val_mean_absolute_error: 0.2360\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 2s 278ms/step - loss: 0.0487 - mean_absolute_error: 0.1538 - val_loss: 0.1476 - val_mean_absolute_error: 0.2364\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 2s 237ms/step - loss: 0.0415 - mean_absolute_error: 0.1410 - val_loss: 0.1597 - val_mean_absolute_error: 0.2424\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 214ms/step - loss: 0.0334 - mean_absolute_error: 0.1232 - val_loss: 0.1441 - val_mean_absolute_error: 0.2167\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.0290 - mean_absolute_error: 0.1133 - val_loss: 0.1398 - val_mean_absolute_error: 0.2115\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.0258 - mean_absolute_error: 0.1056 - val_loss: 0.1519 - val_mean_absolute_error: 0.2182\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0269 - mean_absolute_error: 0.1050 - val_loss: 0.1550 - val_mean_absolute_error: 0.2144\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0204 - mean_absolute_error: 0.0910 - val_loss: 0.1473 - val_mean_absolute_error: 0.2185\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0178 - mean_absolute_error: 0.0859 - val_loss: 0.1359 - val_mean_absolute_error: 0.1984\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0151 - mean_absolute_error: 0.0773 - val_loss: 0.1370 - val_mean_absolute_error: 0.1989\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0164 - mean_absolute_error: 0.0815 - val_loss: 0.1434 - val_mean_absolute_error: 0.1923\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0174 - mean_absolute_error: 0.0813 - val_loss: 0.1499 - val_mean_absolute_error: 0.2023\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0139 - mean_absolute_error: 0.0768 - val_loss: 0.1468 - val_mean_absolute_error: 0.2140\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0154 - mean_absolute_error: 0.0812 - val_loss: 0.1408 - val_mean_absolute_error: 0.2062\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0269 - mean_absolute_error: 0.1109 - val_loss: 0.1455 - val_mean_absolute_error: 0.2183\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 133ms/step - loss: 0.0246 - mean_absolute_error: 0.1047 - val_loss: 0.1570 - val_mean_absolute_error: 0.2199\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0239 - mean_absolute_error: 0.1063 - val_loss: 0.1728 - val_mean_absolute_error: 0.2311\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0188 - mean_absolute_error: 0.0965 - val_loss: 0.1641 - val_mean_absolute_error: 0.2323\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0183 - mean_absolute_error: 0.0990 - val_loss: 0.1546 - val_mean_absolute_error: 0.2176\n",
      "2/2 [==============================] - 0s 72ms/step - loss: 0.1546 - mean_absolute_error: 0.2176\n",
      "2308\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0041 - mean_absolute_error: 0.0436 - val_loss: 0.2032 - val_mean_absolute_error: 0.2563\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0043 - mean_absolute_error: 0.0462 - val_loss: 0.2044 - val_mean_absolute_error: 0.2425\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0037 - mean_absolute_error: 0.0407 - val_loss: 0.2044 - val_mean_absolute_error: 0.2622\n",
      "2/2 [==============================] - 0s 67ms/step - loss: 0.2044 - mean_absolute_error: 0.2622\n",
      "2308\n",
      "Done with seed: 29\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 413ms/step - loss: 0.8160 - mean_absolute_error: 0.7093 - val_loss: 0.5301 - val_mean_absolute_error: 0.5644\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.4820 - mean_absolute_error: 0.5254 - val_loss: 0.4692 - val_mean_absolute_error: 0.5145\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.2882 - mean_absolute_error: 0.3979 - val_loss: 0.4251 - val_mean_absolute_error: 0.4736\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 226ms/step - loss: 0.2377 - mean_absolute_error: 0.3684 - val_loss: 0.3840 - val_mean_absolute_error: 0.4475\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 2s 228ms/step - loss: 0.1814 - mean_absolute_error: 0.2969 - val_loss: 0.3555 - val_mean_absolute_error: 0.3932\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.1334 - mean_absolute_error: 0.2485 - val_loss: 0.3582 - val_mean_absolute_error: 0.3705\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.1056 - mean_absolute_error: 0.2224 - val_loss: 0.3345 - val_mean_absolute_error: 0.3548\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0856 - mean_absolute_error: 0.1975 - val_loss: 0.3068 - val_mean_absolute_error: 0.3332\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0653 - mean_absolute_error: 0.1667 - val_loss: 0.3000 - val_mean_absolute_error: 0.3207\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 188ms/step - loss: 0.0548 - mean_absolute_error: 0.1493 - val_loss: 0.2938 - val_mean_absolute_error: 0.3054\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0441 - mean_absolute_error: 0.1271 - val_loss: 0.2736 - val_mean_absolute_error: 0.2799\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0346 - mean_absolute_error: 0.1121 - val_loss: 0.2925 - val_mean_absolute_error: 0.2775\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 206ms/step - loss: 0.0316 - mean_absolute_error: 0.1056 - val_loss: 0.2839 - val_mean_absolute_error: 0.2654\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 203ms/step - loss: 0.0261 - mean_absolute_error: 0.0959 - val_loss: 0.2864 - val_mean_absolute_error: 0.2695\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 198ms/step - loss: 0.0218 - mean_absolute_error: 0.0858 - val_loss: 0.2897 - val_mean_absolute_error: 0.2713\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 2s 232ms/step - loss: 0.0175 - mean_absolute_error: 0.0740 - val_loss: 0.2777 - val_mean_absolute_error: 0.2504\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 2s 249ms/step - loss: 0.0154 - mean_absolute_error: 0.0704 - val_loss: 0.2770 - val_mean_absolute_error: 0.2618\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 2s 239ms/step - loss: 0.0148 - mean_absolute_error: 0.0636 - val_loss: 0.2797 - val_mean_absolute_error: 0.2582\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 193ms/step - loss: 0.0177 - mean_absolute_error: 0.0774 - val_loss: 0.3089 - val_mean_absolute_error: 0.2843\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 192ms/step - loss: 0.0190 - mean_absolute_error: 0.0826 - val_loss: 0.3153 - val_mean_absolute_error: 0.3141\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.0279 - mean_absolute_error: 0.1026 - val_loss: 0.2770 - val_mean_absolute_error: 0.2872\n",
      "2/2 [==============================] - 0s 122ms/step - loss: 0.2770 - mean_absolute_error: 0.2872\n",
      "2308\n",
      "Done with seed: 30\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 209ms/step - loss: 0.7068 - mean_absolute_error: 0.6776 - val_loss: 0.3510 - val_mean_absolute_error: 0.4957\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.3562 - mean_absolute_error: 0.4581 - val_loss: 0.3241 - val_mean_absolute_error: 0.4473\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.2535 - mean_absolute_error: 0.3783 - val_loss: 0.3121 - val_mean_absolute_error: 0.4161\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.2123 - mean_absolute_error: 0.3398 - val_loss: 0.2417 - val_mean_absolute_error: 0.3582\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.1641 - mean_absolute_error: 0.2941 - val_loss: 0.1915 - val_mean_absolute_error: 0.3128\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.1271 - mean_absolute_error: 0.2588 - val_loss: 0.1552 - val_mean_absolute_error: 0.2642\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0898 - mean_absolute_error: 0.2114 - val_loss: 0.1369 - val_mean_absolute_error: 0.2521\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0718 - mean_absolute_error: 0.1873 - val_loss: 0.1243 - val_mean_absolute_error: 0.2268\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0525 - mean_absolute_error: 0.1589 - val_loss: 0.1184 - val_mean_absolute_error: 0.2253\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0564 - mean_absolute_error: 0.1556 - val_loss: 0.0928 - val_mean_absolute_error: 0.1928\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0473 - mean_absolute_error: 0.1482 - val_loss: 0.0882 - val_mean_absolute_error: 0.1831\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0326 - mean_absolute_error: 0.1260 - val_loss: 0.1164 - val_mean_absolute_error: 0.2183\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0371 - mean_absolute_error: 0.1316 - val_loss: 0.1097 - val_mean_absolute_error: 0.1911\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0269 - mean_absolute_error: 0.1114 - val_loss: 0.0871 - val_mean_absolute_error: 0.1887\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0222 - mean_absolute_error: 0.1017 - val_loss: 0.0821 - val_mean_absolute_error: 0.1881\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.0190 - mean_absolute_error: 0.0953 - val_loss: 0.0806 - val_mean_absolute_error: 0.1678\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0192 - mean_absolute_error: 0.0955 - val_loss: 0.1016 - val_mean_absolute_error: 0.1922\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0167 - mean_absolute_error: 0.0890 - val_loss: 0.1138 - val_mean_absolute_error: 0.2165\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 183ms/step - loss: 0.0202 - mean_absolute_error: 0.1000 - val_loss: 0.0994 - val_mean_absolute_error: 0.1675\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0255 - mean_absolute_error: 0.1155 - val_loss: 0.1660 - val_mean_absolute_error: 0.2830\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0505 - mean_absolute_error: 0.1692 - val_loss: 0.1275 - val_mean_absolute_error: 0.2210\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0462 - mean_absolute_error: 0.1521 - val_loss: 0.1212 - val_mean_absolute_error: 0.2390\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0317 - mean_absolute_error: 0.1265 - val_loss: 0.1103 - val_mean_absolute_error: 0.1968\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0210 - mean_absolute_error: 0.1088 - val_loss: 0.1082 - val_mean_absolute_error: 0.2043\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0200 - mean_absolute_error: 0.1026 - val_loss: 0.0950 - val_mean_absolute_error: 0.1791\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0148 - mean_absolute_error: 0.0882 - val_loss: 0.0907 - val_mean_absolute_error: 0.1755\n",
      "2/2 [==============================] - 0s 75ms/step - loss: 0.0907 - mean_absolute_error: 0.1755\n",
      "2308\n",
      "Done with seed: 31\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 264ms/step - loss: 0.8968 - mean_absolute_error: 0.7743 - val_loss: 0.5624 - val_mean_absolute_error: 0.5490\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 233ms/step - loss: 0.4128 - mean_absolute_error: 0.5118 - val_loss: 0.3712 - val_mean_absolute_error: 0.4670\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 2s 250ms/step - loss: 0.3176 - mean_absolute_error: 0.4465 - val_loss: 0.2983 - val_mean_absolute_error: 0.4111\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 219ms/step - loss: 0.1788 - mean_absolute_error: 0.3208 - val_loss: 0.2415 - val_mean_absolute_error: 0.3458\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 219ms/step - loss: 0.1615 - mean_absolute_error: 0.2926 - val_loss: 0.2152 - val_mean_absolute_error: 0.3291\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 2s 231ms/step - loss: 0.1071 - mean_absolute_error: 0.2462 - val_loss: 0.2109 - val_mean_absolute_error: 0.3245\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 2s 239ms/step - loss: 0.0878 - mean_absolute_error: 0.2167 - val_loss: 0.1842 - val_mean_absolute_error: 0.2821\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 2s 226ms/step - loss: 0.0687 - mean_absolute_error: 0.1859 - val_loss: 0.1903 - val_mean_absolute_error: 0.2774\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 2s 244ms/step - loss: 0.0576 - mean_absolute_error: 0.1671 - val_loss: 0.1900 - val_mean_absolute_error: 0.2689\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 2s 225ms/step - loss: 0.0431 - mean_absolute_error: 0.1396 - val_loss: 0.1907 - val_mean_absolute_error: 0.2636\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 227ms/step - loss: 0.0364 - mean_absolute_error: 0.1240 - val_loss: 0.1822 - val_mean_absolute_error: 0.2544\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 222ms/step - loss: 0.0301 - mean_absolute_error: 0.1127 - val_loss: 0.1839 - val_mean_absolute_error: 0.2556\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 2s 240ms/step - loss: 0.0273 - mean_absolute_error: 0.1089 - val_loss: 0.1742 - val_mean_absolute_error: 0.2503\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 2s 290ms/step - loss: 0.0268 - mean_absolute_error: 0.1045 - val_loss: 0.1598 - val_mean_absolute_error: 0.2108\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 2s 264ms/step - loss: 0.0159 - mean_absolute_error: 0.0814 - val_loss: 0.1738 - val_mean_absolute_error: 0.2218\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 2s 301ms/step - loss: 0.0232 - mean_absolute_error: 0.0971 - val_loss: 0.1763 - val_mean_absolute_error: 0.2174\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 2s 313ms/step - loss: 0.0279 - mean_absolute_error: 0.1023 - val_loss: 0.1876 - val_mean_absolute_error: 0.2762\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 2s 330ms/step - loss: 0.0504 - mean_absolute_error: 0.1477 - val_loss: 0.1726 - val_mean_absolute_error: 0.2324\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 3s 384ms/step - loss: 0.0321 - mean_absolute_error: 0.1211 - val_loss: 0.1606 - val_mean_absolute_error: 0.2221\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 2s 336ms/step - loss: 0.0373 - mean_absolute_error: 0.1281 - val_loss: 0.1697 - val_mean_absolute_error: 0.2549\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 2s 276ms/step - loss: 0.0320 - mean_absolute_error: 0.1206 - val_loss: 0.1543 - val_mean_absolute_error: 0.2397\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 202ms/step - loss: 0.0259 - mean_absolute_error: 0.1123 - val_loss: 0.1360 - val_mean_absolute_error: 0.1945\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 210ms/step - loss: 0.0196 - mean_absolute_error: 0.0957 - val_loss: 0.1350 - val_mean_absolute_error: 0.1915\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.0238 - mean_absolute_error: 0.1074 - val_loss: 0.1467 - val_mean_absolute_error: 0.2164\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0164 - mean_absolute_error: 0.0841 - val_loss: 0.1464 - val_mean_absolute_error: 0.2106\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0175 - mean_absolute_error: 0.0893 - val_loss: 0.1337 - val_mean_absolute_error: 0.1818\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0088 - mean_absolute_error: 0.0651 - val_loss: 0.1338 - val_mean_absolute_error: 0.1835\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0120 - mean_absolute_error: 0.0735 - val_loss: 0.1370 - val_mean_absolute_error: 0.1898\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0080 - mean_absolute_error: 0.0582 - val_loss: 0.1376 - val_mean_absolute_error: 0.1866\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0091 - mean_absolute_error: 0.0654 - val_loss: 0.1266 - val_mean_absolute_error: 0.1722\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0067 - mean_absolute_error: 0.0552 - val_loss: 0.1180 - val_mean_absolute_error: 0.1671\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0078 - mean_absolute_error: 0.0615 - val_loss: 0.1272 - val_mean_absolute_error: 0.1799\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 133ms/step - loss: 0.0060 - mean_absolute_error: 0.0529 - val_loss: 0.1298 - val_mean_absolute_error: 0.1728\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 139ms/step - loss: 0.0088 - mean_absolute_error: 0.0635 - val_loss: 0.1194 - val_mean_absolute_error: 0.1746\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0064 - mean_absolute_error: 0.0525 - val_loss: 0.1163 - val_mean_absolute_error: 0.1621\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0054 - mean_absolute_error: 0.0473 - val_loss: 0.1217 - val_mean_absolute_error: 0.1693\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0045 - mean_absolute_error: 0.0418 - val_loss: 0.1221 - val_mean_absolute_error: 0.1620\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0045 - mean_absolute_error: 0.0413 - val_loss: 0.1228 - val_mean_absolute_error: 0.1629\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.0041 - mean_absolute_error: 0.0389 - val_loss: 0.1181 - val_mean_absolute_error: 0.1562\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0044 - mean_absolute_error: 0.0421 - val_loss: 0.1166 - val_mean_absolute_error: 0.1632\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0029 - mean_absolute_error: 0.0334 - val_loss: 0.1195 - val_mean_absolute_error: 0.1593\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0041 - mean_absolute_error: 0.0423 - val_loss: 0.1214 - val_mean_absolute_error: 0.1549\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0034 - mean_absolute_error: 0.0336 - val_loss: 0.1247 - val_mean_absolute_error: 0.1641\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0051 - mean_absolute_error: 0.0457 - val_loss: 0.1213 - val_mean_absolute_error: 0.1536\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0040 - mean_absolute_error: 0.0373 - val_loss: 0.1278 - val_mean_absolute_error: 0.1817\n",
      "2/2 [==============================] - 0s 76ms/step - loss: 0.1278 - mean_absolute_error: 0.1817\n",
      "2308\n",
      "Done with seed: 32\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 433ms/step - loss: 1.1157 - mean_absolute_error: 0.8417 - val_loss: 0.5157 - val_mean_absolute_error: 0.5849\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.4466 - mean_absolute_error: 0.5330 - val_loss: 0.4386 - val_mean_absolute_error: 0.5115\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.3209 - mean_absolute_error: 0.4464 - val_loss: 0.3842 - val_mean_absolute_error: 0.4620\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.2509 - mean_absolute_error: 0.3777 - val_loss: 0.3395 - val_mean_absolute_error: 0.4157\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.1855 - mean_absolute_error: 0.3183 - val_loss: 0.2727 - val_mean_absolute_error: 0.3584\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.1368 - mean_absolute_error: 0.2661 - val_loss: 0.2504 - val_mean_absolute_error: 0.3318\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 174ms/step - loss: 0.1047 - mean_absolute_error: 0.2298 - val_loss: 0.2441 - val_mean_absolute_error: 0.3330\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.0807 - mean_absolute_error: 0.1998 - val_loss: 0.2303 - val_mean_absolute_error: 0.3080\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0592 - mean_absolute_error: 0.1669 - val_loss: 0.2072 - val_mean_absolute_error: 0.2784\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 197ms/step - loss: 0.0440 - mean_absolute_error: 0.1382 - val_loss: 0.1986 - val_mean_absolute_error: 0.2575\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.0327 - mean_absolute_error: 0.1163 - val_loss: 0.1910 - val_mean_absolute_error: 0.2495\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 196ms/step - loss: 0.0267 - mean_absolute_error: 0.1022 - val_loss: 0.1777 - val_mean_absolute_error: 0.2315\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 192ms/step - loss: 0.0197 - mean_absolute_error: 0.0894 - val_loss: 0.1757 - val_mean_absolute_error: 0.2223\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 191ms/step - loss: 0.0159 - mean_absolute_error: 0.0790 - val_loss: 0.1638 - val_mean_absolute_error: 0.2144\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0117 - mean_absolute_error: 0.0724 - val_loss: 0.1624 - val_mean_absolute_error: 0.2068\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 224ms/step - loss: 0.0102 - mean_absolute_error: 0.0675 - val_loss: 0.1681 - val_mean_absolute_error: 0.2105\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 216ms/step - loss: 0.0087 - mean_absolute_error: 0.0611 - val_loss: 0.1676 - val_mean_absolute_error: 0.2255\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 220ms/step - loss: 0.0106 - mean_absolute_error: 0.0674 - val_loss: 0.1892 - val_mean_absolute_error: 0.2584\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 222ms/step - loss: 0.0172 - mean_absolute_error: 0.0844 - val_loss: 0.1850 - val_mean_absolute_error: 0.2394\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 2s 286ms/step - loss: 0.0206 - mean_absolute_error: 0.0954 - val_loss: 0.1747 - val_mean_absolute_error: 0.2307\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 2s 319ms/step - loss: 0.0313 - mean_absolute_error: 0.1201 - val_loss: 0.2403 - val_mean_absolute_error: 0.2944\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 2s 375ms/step - loss: 0.0475 - mean_absolute_error: 0.1534 - val_loss: 0.1721 - val_mean_absolute_error: 0.2588\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 2s 347ms/step - loss: 0.0392 - mean_absolute_error: 0.1422 - val_loss: 0.2076 - val_mean_absolute_error: 0.2899\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.0350 - mean_absolute_error: 0.1337 - val_loss: 0.1700 - val_mean_absolute_error: 0.2387\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.0185 - mean_absolute_error: 0.0991 - val_loss: 0.1803 - val_mean_absolute_error: 0.2309\n",
      "2/2 [==============================] - 0s 100ms/step - loss: 0.1803 - mean_absolute_error: 0.2309\n",
      "2308\n",
      "Done with seed: 33\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 2s 185ms/step - loss: 1.1327 - mean_absolute_error: 0.8430 - val_loss: 0.6070 - val_mean_absolute_error: 0.5867\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.4466 - mean_absolute_error: 0.5163 - val_loss: 0.3727 - val_mean_absolute_error: 0.4796\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.2776 - mean_absolute_error: 0.4073 - val_loss: 0.3153 - val_mean_absolute_error: 0.4272\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.1936 - mean_absolute_error: 0.3314 - val_loss: 0.3310 - val_mean_absolute_error: 0.3895\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.1386 - mean_absolute_error: 0.2639 - val_loss: 0.2677 - val_mean_absolute_error: 0.3446\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.1030 - mean_absolute_error: 0.2249 - val_loss: 0.2585 - val_mean_absolute_error: 0.3215\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0772 - mean_absolute_error: 0.1895 - val_loss: 0.2612 - val_mean_absolute_error: 0.3097\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0521 - mean_absolute_error: 0.1536 - val_loss: 0.2515 - val_mean_absolute_error: 0.2949\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0374 - mean_absolute_error: 0.1263 - val_loss: 0.2469 - val_mean_absolute_error: 0.2808\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0298 - mean_absolute_error: 0.1078 - val_loss: 0.2386 - val_mean_absolute_error: 0.2690\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0249 - mean_absolute_error: 0.0989 - val_loss: 0.2399 - val_mean_absolute_error: 0.2767\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0219 - mean_absolute_error: 0.0948 - val_loss: 0.2478 - val_mean_absolute_error: 0.2853\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0213 - mean_absolute_error: 0.0931 - val_loss: 0.2301 - val_mean_absolute_error: 0.2478\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0178 - mean_absolute_error: 0.0873 - val_loss: 0.2120 - val_mean_absolute_error: 0.2606\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0198 - mean_absolute_error: 0.0973 - val_loss: 0.2165 - val_mean_absolute_error: 0.2745\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0232 - mean_absolute_error: 0.1048 - val_loss: 0.2105 - val_mean_absolute_error: 0.2492\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.0244 - mean_absolute_error: 0.1017 - val_loss: 0.2203 - val_mean_absolute_error: 0.2705\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0236 - mean_absolute_error: 0.1072 - val_loss: 0.2547 - val_mean_absolute_error: 0.3350\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0409 - mean_absolute_error: 0.1429 - val_loss: 0.2089 - val_mean_absolute_error: 0.2451\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0325 - mean_absolute_error: 0.1280 - val_loss: 0.2413 - val_mean_absolute_error: 0.3104\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0403 - mean_absolute_error: 0.1466 - val_loss: 0.2152 - val_mean_absolute_error: 0.2456\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0244 - mean_absolute_error: 0.1118 - val_loss: 0.2272 - val_mean_absolute_error: 0.2886\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0198 - mean_absolute_error: 0.0992 - val_loss: 0.2091 - val_mean_absolute_error: 0.2294\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0120 - mean_absolute_error: 0.0762 - val_loss: 0.2153 - val_mean_absolute_error: 0.2572\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0141 - mean_absolute_error: 0.0797 - val_loss: 0.2116 - val_mean_absolute_error: 0.2497\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0114 - mean_absolute_error: 0.0723 - val_loss: 0.2082 - val_mean_absolute_error: 0.2548\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0098 - mean_absolute_error: 0.0686 - val_loss: 0.2090 - val_mean_absolute_error: 0.2394\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0096 - mean_absolute_error: 0.0578 - val_loss: 0.2031 - val_mean_absolute_error: 0.2326\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0062 - mean_absolute_error: 0.0530 - val_loss: 0.2060 - val_mean_absolute_error: 0.2246\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0048 - mean_absolute_error: 0.0446 - val_loss: 0.2034 - val_mean_absolute_error: 0.2244\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0046 - mean_absolute_error: 0.0402 - val_loss: 0.2008 - val_mean_absolute_error: 0.2080\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0032 - mean_absolute_error: 0.0316 - val_loss: 0.1999 - val_mean_absolute_error: 0.2149\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0032 - mean_absolute_error: 0.0301 - val_loss: 0.1999 - val_mean_absolute_error: 0.2068\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.0030 - mean_absolute_error: 0.0274 - val_loss: 0.2013 - val_mean_absolute_error: 0.2148\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0031 - mean_absolute_error: 0.0285 - val_loss: 0.2030 - val_mean_absolute_error: 0.2163\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0035 - mean_absolute_error: 0.0327 - val_loss: 0.2000 - val_mean_absolute_error: 0.2061\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 213ms/step - loss: 0.0030 - mean_absolute_error: 0.0284 - val_loss: 0.2003 - val_mean_absolute_error: 0.2222\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 193ms/step - loss: 0.0041 - mean_absolute_error: 0.0381 - val_loss: 0.1996 - val_mean_absolute_error: 0.2131\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0043 - mean_absolute_error: 0.0394 - val_loss: 0.2046 - val_mean_absolute_error: 0.2248\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 201ms/step - loss: 0.0046 - mean_absolute_error: 0.0424 - val_loss: 0.2101 - val_mean_absolute_error: 0.2387\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 206ms/step - loss: 0.0077 - mean_absolute_error: 0.0589 - val_loss: 0.2036 - val_mean_absolute_error: 0.2112\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 183ms/step - loss: 0.0060 - mean_absolute_error: 0.0510 - val_loss: 0.2073 - val_mean_absolute_error: 0.2500\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 188ms/step - loss: 0.0101 - mean_absolute_error: 0.0702 - val_loss: 0.2037 - val_mean_absolute_error: 0.2288\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 193ms/step - loss: 0.0103 - mean_absolute_error: 0.0695 - val_loss: 0.2139 - val_mean_absolute_error: 0.2543\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0103 - mean_absolute_error: 0.0718 - val_loss: 0.2171 - val_mean_absolute_error: 0.2532\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0143 - mean_absolute_error: 0.0858 - val_loss: 0.2113 - val_mean_absolute_error: 0.2476\n",
      "Epoch 47/1000\n",
      "7/7 [==============================] - 1s 184ms/step - loss: 0.0104 - mean_absolute_error: 0.0716 - val_loss: 0.2114 - val_mean_absolute_error: 0.2503\n",
      "Epoch 48/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0133 - mean_absolute_error: 0.0849 - val_loss: 0.2124 - val_mean_absolute_error: 0.2479\n",
      "2/2 [==============================] - 0s 74ms/step - loss: 0.2124 - mean_absolute_error: 0.2479\n",
      "2308\n",
      "Done with seed: 34\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 241ms/step - loss: 0.8937 - mean_absolute_error: 0.7564 - val_loss: 0.5714 - val_mean_absolute_error: 0.5622\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.6364 - mean_absolute_error: 0.6429 - val_loss: 0.4736 - val_mean_absolute_error: 0.5360\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.2626 - mean_absolute_error: 0.3956 - val_loss: 0.2781 - val_mean_absolute_error: 0.3886\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.2317 - mean_absolute_error: 0.3739 - val_loss: 0.2765 - val_mean_absolute_error: 0.3784\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.1532 - mean_absolute_error: 0.2864 - val_loss: 0.2379 - val_mean_absolute_error: 0.3382\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.1187 - mean_absolute_error: 0.2351 - val_loss: 0.2006 - val_mean_absolute_error: 0.2926\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0932 - mean_absolute_error: 0.1985 - val_loss: 0.1783 - val_mean_absolute_error: 0.2621\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0709 - mean_absolute_error: 0.1710 - val_loss: 0.1722 - val_mean_absolute_error: 0.2619\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0547 - mean_absolute_error: 0.1499 - val_loss: 0.1569 - val_mean_absolute_error: 0.2461\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 160ms/step - loss: 0.0454 - mean_absolute_error: 0.1376 - val_loss: 0.1621 - val_mean_absolute_error: 0.2399\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 155ms/step - loss: 0.0286 - mean_absolute_error: 0.1123 - val_loss: 0.1861 - val_mean_absolute_error: 0.2590\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 152ms/step - loss: 0.0196 - mean_absolute_error: 0.0948 - val_loss: 0.1836 - val_mean_absolute_error: 0.2475\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0162 - mean_absolute_error: 0.0858 - val_loss: 0.1575 - val_mean_absolute_error: 0.2150\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0176 - mean_absolute_error: 0.0851 - val_loss: 0.1426 - val_mean_absolute_error: 0.2182\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 146ms/step - loss: 0.0180 - mean_absolute_error: 0.0892 - val_loss: 0.1617 - val_mean_absolute_error: 0.2275\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0148 - mean_absolute_error: 0.0843 - val_loss: 0.1682 - val_mean_absolute_error: 0.2394\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 135ms/step - loss: 0.0119 - mean_absolute_error: 0.0774 - val_loss: 0.1474 - val_mean_absolute_error: 0.2131\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0069 - mean_absolute_error: 0.0611 - val_loss: 0.1401 - val_mean_absolute_error: 0.2022\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0062 - mean_absolute_error: 0.0578 - val_loss: 0.1324 - val_mean_absolute_error: 0.1904\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0056 - mean_absolute_error: 0.0539 - val_loss: 0.1453 - val_mean_absolute_error: 0.2053\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0052 - mean_absolute_error: 0.0511 - val_loss: 0.1845 - val_mean_absolute_error: 0.2643\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0122 - mean_absolute_error: 0.0718 - val_loss: 0.1671 - val_mean_absolute_error: 0.2242\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0133 - mean_absolute_error: 0.0799 - val_loss: 0.1470 - val_mean_absolute_error: 0.2277\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0251 - mean_absolute_error: 0.1127 - val_loss: 0.1435 - val_mean_absolute_error: 0.2149\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0210 - mean_absolute_error: 0.1049 - val_loss: 0.2376 - val_mean_absolute_error: 0.3234\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.0288 - mean_absolute_error: 0.1227 - val_loss: 0.1624 - val_mean_absolute_error: 0.2280\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 140ms/step - loss: 0.0157 - mean_absolute_error: 0.0948 - val_loss: 0.1647 - val_mean_absolute_error: 0.2403\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0189 - mean_absolute_error: 0.0984 - val_loss: 0.1869 - val_mean_absolute_error: 0.2557\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 149ms/step - loss: 0.0105 - mean_absolute_error: 0.0770 - val_loss: 0.1598 - val_mean_absolute_error: 0.2035\n",
      "2/2 [==============================] - 0s 74ms/step - loss: 0.1598 - mean_absolute_error: 0.2035\n",
      "2308\n",
      "Done with seed: 35\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 239ms/step - loss: 1.0537 - mean_absolute_error: 0.8330 - val_loss: 0.5192 - val_mean_absolute_error: 0.6166\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.4545 - mean_absolute_error: 0.5181 - val_loss: 0.3338 - val_mean_absolute_error: 0.4799\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.2918 - mean_absolute_error: 0.4215 - val_loss: 0.2933 - val_mean_absolute_error: 0.4257\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.2449 - mean_absolute_error: 0.3715 - val_loss: 0.2493 - val_mean_absolute_error: 0.3764\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 175ms/step - loss: 0.1977 - mean_absolute_error: 0.3133 - val_loss: 0.1966 - val_mean_absolute_error: 0.3428\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 187ms/step - loss: 0.1641 - mean_absolute_error: 0.2995 - val_loss: 0.1928 - val_mean_absolute_error: 0.3224\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.1354 - mean_absolute_error: 0.2592 - val_loss: 0.1588 - val_mean_absolute_error: 0.2838\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 217ms/step - loss: 0.1066 - mean_absolute_error: 0.2255 - val_loss: 0.1518 - val_mean_absolute_error: 0.2755\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 2s 259ms/step - loss: 0.0937 - mean_absolute_error: 0.2132 - val_loss: 0.1645 - val_mean_absolute_error: 0.2889\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 2s 286ms/step - loss: 0.0854 - mean_absolute_error: 0.1983 - val_loss: 0.1430 - val_mean_absolute_error: 0.2596\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 2s 345ms/step - loss: 0.0634 - mean_absolute_error: 0.1693 - val_loss: 0.1352 - val_mean_absolute_error: 0.2552\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 2s 355ms/step - loss: 0.0553 - mean_absolute_error: 0.1531 - val_loss: 0.1174 - val_mean_absolute_error: 0.2151\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 2s 346ms/step - loss: 0.0412 - mean_absolute_error: 0.1244 - val_loss: 0.1159 - val_mean_absolute_error: 0.1993\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 2s 287ms/step - loss: 0.0336 - mean_absolute_error: 0.1108 - val_loss: 0.1176 - val_mean_absolute_error: 0.1915\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 2s 261ms/step - loss: 0.0332 - mean_absolute_error: 0.1003 - val_loss: 0.1064 - val_mean_absolute_error: 0.1773\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 2s 281ms/step - loss: 0.0277 - mean_absolute_error: 0.0890 - val_loss: 0.1040 - val_mean_absolute_error: 0.1632\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 2s 228ms/step - loss: 0.0231 - mean_absolute_error: 0.0774 - val_loss: 0.1040 - val_mean_absolute_error: 0.1639\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 216ms/step - loss: 0.0207 - mean_absolute_error: 0.0675 - val_loss: 0.1004 - val_mean_absolute_error: 0.1558\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 2s 233ms/step - loss: 0.0193 - mean_absolute_error: 0.0637 - val_loss: 0.0951 - val_mean_absolute_error: 0.1479\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 2s 230ms/step - loss: 0.0222 - mean_absolute_error: 0.0669 - val_loss: 0.1048 - val_mean_absolute_error: 0.1760\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.0268 - mean_absolute_error: 0.0842 - val_loss: 0.0955 - val_mean_absolute_error: 0.1714\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 176ms/step - loss: 0.0240 - mean_absolute_error: 0.0809 - val_loss: 0.0972 - val_mean_absolute_error: 0.1676\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0175 - mean_absolute_error: 0.0716 - val_loss: 0.1018 - val_mean_absolute_error: 0.1635\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0159 - mean_absolute_error: 0.0688 - val_loss: 0.0995 - val_mean_absolute_error: 0.1513\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0153 - mean_absolute_error: 0.0678 - val_loss: 0.1112 - val_mean_absolute_error: 0.1668\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.0162 - mean_absolute_error: 0.0716 - val_loss: 0.1176 - val_mean_absolute_error: 0.1861\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0157 - mean_absolute_error: 0.0772 - val_loss: 0.1081 - val_mean_absolute_error: 0.1544\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0156 - mean_absolute_error: 0.0739 - val_loss: 0.1184 - val_mean_absolute_error: 0.2004\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0193 - mean_absolute_error: 0.0911 - val_loss: 0.1033 - val_mean_absolute_error: 0.1511\n",
      "2/2 [==============================] - 0s 68ms/step - loss: 0.1033 - mean_absolute_error: 0.1511\n",
      "2308\n",
      "Done with seed: 36\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 227ms/step - loss: 0.8213 - mean_absolute_error: 0.7280 - val_loss: 0.4308 - val_mean_absolute_error: 0.5286\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.4482 - mean_absolute_error: 0.5398 - val_loss: 0.3199 - val_mean_absolute_error: 0.4404\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.2752 - mean_absolute_error: 0.3967 - val_loss: 0.3171 - val_mean_absolute_error: 0.4254\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.2896 - mean_absolute_error: 0.4197 - val_loss: 0.3363 - val_mean_absolute_error: 0.4348\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 141ms/step - loss: 0.2298 - mean_absolute_error: 0.3579 - val_loss: 0.2330 - val_mean_absolute_error: 0.3713\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.1415 - mean_absolute_error: 0.2878 - val_loss: 0.1874 - val_mean_absolute_error: 0.3088\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.1144 - mean_absolute_error: 0.2451 - val_loss: 0.1353 - val_mean_absolute_error: 0.2715\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0731 - mean_absolute_error: 0.2002 - val_loss: 0.1515 - val_mean_absolute_error: 0.2597\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0554 - mean_absolute_error: 0.1697 - val_loss: 0.1077 - val_mean_absolute_error: 0.2215\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0392 - mean_absolute_error: 0.1377 - val_loss: 0.0955 - val_mean_absolute_error: 0.1930\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 156ms/step - loss: 0.0330 - mean_absolute_error: 0.1209 - val_loss: 0.0944 - val_mean_absolute_error: 0.1861\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0281 - mean_absolute_error: 0.1125 - val_loss: 0.0856 - val_mean_absolute_error: 0.1725\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0231 - mean_absolute_error: 0.0984 - val_loss: 0.0781 - val_mean_absolute_error: 0.1560\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 186ms/step - loss: 0.0170 - mean_absolute_error: 0.0836 - val_loss: 0.0742 - val_mean_absolute_error: 0.1480\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0114 - mean_absolute_error: 0.0662 - val_loss: 0.0697 - val_mean_absolute_error: 0.1325\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 147ms/step - loss: 0.0088 - mean_absolute_error: 0.0583 - val_loss: 0.0691 - val_mean_absolute_error: 0.1318\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0088 - mean_absolute_error: 0.0559 - val_loss: 0.0734 - val_mean_absolute_error: 0.1430\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0105 - mean_absolute_error: 0.0617 - val_loss: 0.0758 - val_mean_absolute_error: 0.1431\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0101 - mean_absolute_error: 0.0657 - val_loss: 0.0680 - val_mean_absolute_error: 0.1250\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 150ms/step - loss: 0.0103 - mean_absolute_error: 0.0653 - val_loss: 0.0722 - val_mean_absolute_error: 0.1573\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0153 - mean_absolute_error: 0.0841 - val_loss: 0.0652 - val_mean_absolute_error: 0.1341\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 151ms/step - loss: 0.0212 - mean_absolute_error: 0.1021 - val_loss: 0.0865 - val_mean_absolute_error: 0.1913\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0226 - mean_absolute_error: 0.1132 - val_loss: 0.0707 - val_mean_absolute_error: 0.1419\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0175 - mean_absolute_error: 0.0953 - val_loss: 0.0762 - val_mean_absolute_error: 0.1711\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0193 - mean_absolute_error: 0.1012 - val_loss: 0.0640 - val_mean_absolute_error: 0.1290\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 154ms/step - loss: 0.0128 - mean_absolute_error: 0.0819 - val_loss: 0.0787 - val_mean_absolute_error: 0.1674\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0174 - mean_absolute_error: 0.0932 - val_loss: 0.0658 - val_mean_absolute_error: 0.1425\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0134 - mean_absolute_error: 0.0817 - val_loss: 0.0624 - val_mean_absolute_error: 0.1305\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 162ms/step - loss: 0.0105 - mean_absolute_error: 0.0704 - val_loss: 0.0703 - val_mean_absolute_error: 0.1455\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0079 - mean_absolute_error: 0.0637 - val_loss: 0.0621 - val_mean_absolute_error: 0.1096\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0063 - mean_absolute_error: 0.0499 - val_loss: 0.0586 - val_mean_absolute_error: 0.1181\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 163ms/step - loss: 0.0054 - mean_absolute_error: 0.0458 - val_loss: 0.0587 - val_mean_absolute_error: 0.1080\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0042 - mean_absolute_error: 0.0384 - val_loss: 0.0615 - val_mean_absolute_error: 0.1114\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0040 - mean_absolute_error: 0.0372 - val_loss: 0.0586 - val_mean_absolute_error: 0.1064\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 158ms/step - loss: 0.0036 - mean_absolute_error: 0.0333 - val_loss: 0.0562 - val_mean_absolute_error: 0.0981\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 157ms/step - loss: 0.0032 - mean_absolute_error: 0.0292 - val_loss: 0.0609 - val_mean_absolute_error: 0.1081\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0032 - mean_absolute_error: 0.0288 - val_loss: 0.0597 - val_mean_absolute_error: 0.0977\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 210ms/step - loss: 0.0046 - mean_absolute_error: 0.0265 - val_loss: 0.0599 - val_mean_absolute_error: 0.1040\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 194ms/step - loss: 0.0033 - mean_absolute_error: 0.0264 - val_loss: 0.0607 - val_mean_absolute_error: 0.1029\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 2s 233ms/step - loss: 0.0028 - mean_absolute_error: 0.0239 - val_loss: 0.0607 - val_mean_absolute_error: 0.1010\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0024 - mean_absolute_error: 0.0199 - val_loss: 0.0608 - val_mean_absolute_error: 0.0982\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 223ms/step - loss: 0.0025 - mean_absolute_error: 0.0198 - val_loss: 0.0596 - val_mean_absolute_error: 0.0961\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 195ms/step - loss: 0.0024 - mean_absolute_error: 0.0197 - val_loss: 0.0592 - val_mean_absolute_error: 0.0988\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 203ms/step - loss: 0.0023 - mean_absolute_error: 0.0219 - val_loss: 0.0612 - val_mean_absolute_error: 0.1024\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0028 - mean_absolute_error: 0.0262 - val_loss: 0.0628 - val_mean_absolute_error: 0.1128\n",
      "2/2 [==============================] - 0s 83ms/step - loss: 0.0628 - mean_absolute_error: 0.1128\n",
      "2308\n",
      "Done with seed: 37\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 3s 275ms/step - loss: 0.8355 - mean_absolute_error: 0.7252 - val_loss: 0.6211 - val_mean_absolute_error: 0.6344\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 2s 248ms/step - loss: 0.4384 - mean_absolute_error: 0.5279 - val_loss: 0.5209 - val_mean_absolute_error: 0.5034\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 221ms/step - loss: 0.2446 - mean_absolute_error: 0.3684 - val_loss: 0.3814 - val_mean_absolute_error: 0.4373\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.1675 - mean_absolute_error: 0.3055 - val_loss: 0.3056 - val_mean_absolute_error: 0.3793\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 203ms/step - loss: 0.1174 - mean_absolute_error: 0.2434 - val_loss: 0.2438 - val_mean_absolute_error: 0.3202\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0848 - mean_absolute_error: 0.2050 - val_loss: 0.2251 - val_mean_absolute_error: 0.3039\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0700 - mean_absolute_error: 0.1809 - val_loss: 0.2265 - val_mean_absolute_error: 0.2878\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.0428 - mean_absolute_error: 0.1425 - val_loss: 0.2309 - val_mean_absolute_error: 0.2988\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 191ms/step - loss: 0.0432 - mean_absolute_error: 0.1323 - val_loss: 0.2037 - val_mean_absolute_error: 0.2704\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0412 - mean_absolute_error: 0.1381 - val_loss: 0.1806 - val_mean_absolute_error: 0.2532\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.0301 - mean_absolute_error: 0.1181 - val_loss: 0.1733 - val_mean_absolute_error: 0.2478\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0294 - mean_absolute_error: 0.1180 - val_loss: 0.2054 - val_mean_absolute_error: 0.2390\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 145ms/step - loss: 0.0214 - mean_absolute_error: 0.0998 - val_loss: 0.2526 - val_mean_absolute_error: 0.2952\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0276 - mean_absolute_error: 0.1121 - val_loss: 0.2204 - val_mean_absolute_error: 0.2351\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 144ms/step - loss: 0.0172 - mean_absolute_error: 0.0862 - val_loss: 0.2048 - val_mean_absolute_error: 0.2688\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 143ms/step - loss: 0.0161 - mean_absolute_error: 0.0872 - val_loss: 0.1990 - val_mean_absolute_error: 0.2297\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0120 - mean_absolute_error: 0.0758 - val_loss: 0.2480 - val_mean_absolute_error: 0.2633\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 148ms/step - loss: 0.0099 - mean_absolute_error: 0.0682 - val_loss: 0.2426 - val_mean_absolute_error: 0.2552\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 153ms/step - loss: 0.0113 - mean_absolute_error: 0.0699 - val_loss: 0.1835 - val_mean_absolute_error: 0.2177\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.0104 - mean_absolute_error: 0.0692 - val_loss: 0.1884 - val_mean_absolute_error: 0.2537\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.0158 - mean_absolute_error: 0.0862 - val_loss: 0.2600 - val_mean_absolute_error: 0.2755\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2/2 [==============================] - 0s 66ms/step - loss: 0.2600 - mean_absolute_error: 0.2755\n",
      "2308\n",
      "Done with seed: 38\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 4s 219ms/step - loss: 1.0228 - mean_absolute_error: 0.7984 - val_loss: 0.7605 - val_mean_absolute_error: 0.7536\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 177ms/step - loss: 0.5976 - mean_absolute_error: 0.6284 - val_loss: 0.5476 - val_mean_absolute_error: 0.5752\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.3824 - mean_absolute_error: 0.4776 - val_loss: 0.3870 - val_mean_absolute_error: 0.5171\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 161ms/step - loss: 0.2931 - mean_absolute_error: 0.4156 - val_loss: 0.2905 - val_mean_absolute_error: 0.4235\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 166ms/step - loss: 0.2271 - mean_absolute_error: 0.3460 - val_loss: 0.2391 - val_mean_absolute_error: 0.3883\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 167ms/step - loss: 0.1567 - mean_absolute_error: 0.2908 - val_loss: 0.2209 - val_mean_absolute_error: 0.3721\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 159ms/step - loss: 0.1312 - mean_absolute_error: 0.2594 - val_loss: 0.1860 - val_mean_absolute_error: 0.3348\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 164ms/step - loss: 0.1126 - mean_absolute_error: 0.2415 - val_loss: 0.2002 - val_mean_absolute_error: 0.3249\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0916 - mean_absolute_error: 0.2039 - val_loss: 0.1735 - val_mean_absolute_error: 0.3171\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 172ms/step - loss: 0.0787 - mean_absolute_error: 0.2008 - val_loss: 0.1781 - val_mean_absolute_error: 0.2862\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 171ms/step - loss: 0.0728 - mean_absolute_error: 0.1829 - val_loss: 0.1650 - val_mean_absolute_error: 0.2727\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0620 - mean_absolute_error: 0.1689 - val_loss: 0.1586 - val_mean_absolute_error: 0.2574\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 214ms/step - loss: 0.0488 - mean_absolute_error: 0.1475 - val_loss: 0.1530 - val_mean_absolute_error: 0.2562\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 2s 227ms/step - loss: 0.0443 - mean_absolute_error: 0.1378 - val_loss: 0.1449 - val_mean_absolute_error: 0.2520\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 2s 245ms/step - loss: 0.0368 - mean_absolute_error: 0.1326 - val_loss: 0.1450 - val_mean_absolute_error: 0.2485\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 2s 247ms/step - loss: 0.0269 - mean_absolute_error: 0.1130 - val_loss: 0.1281 - val_mean_absolute_error: 0.2192\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 189ms/step - loss: 0.0182 - mean_absolute_error: 0.0891 - val_loss: 0.1238 - val_mean_absolute_error: 0.2095\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 193ms/step - loss: 0.0133 - mean_absolute_error: 0.0745 - val_loss: 0.1202 - val_mean_absolute_error: 0.1988\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 181ms/step - loss: 0.0099 - mean_absolute_error: 0.0621 - val_loss: 0.1179 - val_mean_absolute_error: 0.1853\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 168ms/step - loss: 0.0070 - mean_absolute_error: 0.0518 - val_loss: 0.1187 - val_mean_absolute_error: 0.1825\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0054 - mean_absolute_error: 0.0448 - val_loss: 0.1146 - val_mean_absolute_error: 0.1743\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 179ms/step - loss: 0.0043 - mean_absolute_error: 0.0377 - val_loss: 0.1132 - val_mean_absolute_error: 0.1679\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 211ms/step - loss: 0.0034 - mean_absolute_error: 0.0339 - val_loss: 0.1136 - val_mean_absolute_error: 0.1670\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 173ms/step - loss: 0.0028 - mean_absolute_error: 0.0307 - val_loss: 0.1148 - val_mean_absolute_error: 0.1692\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0023 - mean_absolute_error: 0.0280 - val_loss: 0.1097 - val_mean_absolute_error: 0.1638\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0022 - mean_absolute_error: 0.0280 - val_loss: 0.1013 - val_mean_absolute_error: 0.1633\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 185ms/step - loss: 0.0034 - mean_absolute_error: 0.0317 - val_loss: 0.1136 - val_mean_absolute_error: 0.1925\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 212ms/step - loss: 0.0066 - mean_absolute_error: 0.0478 - val_loss: 0.1002 - val_mean_absolute_error: 0.1732\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 222ms/step - loss: 0.0051 - mean_absolute_error: 0.0463 - val_loss: 0.1013 - val_mean_absolute_error: 0.1739\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 222ms/step - loss: 0.0077 - mean_absolute_error: 0.0581 - val_loss: 0.1059 - val_mean_absolute_error: 0.1874\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 2s 250ms/step - loss: 0.0074 - mean_absolute_error: 0.0576 - val_loss: 0.1029 - val_mean_absolute_error: 0.1749\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 209ms/step - loss: 0.0067 - mean_absolute_error: 0.0572 - val_loss: 0.1032 - val_mean_absolute_error: 0.1753\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 204ms/step - loss: 0.0056 - mean_absolute_error: 0.0550 - val_loss: 0.1166 - val_mean_absolute_error: 0.2115\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 210ms/step - loss: 0.0111 - mean_absolute_error: 0.0762 - val_loss: 0.1138 - val_mean_absolute_error: 0.1825\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.0130 - mean_absolute_error: 0.0801 - val_loss: 0.1526 - val_mean_absolute_error: 0.2474\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 200ms/step - loss: 0.0258 - mean_absolute_error: 0.1149 - val_loss: 0.0998 - val_mean_absolute_error: 0.1853\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 201ms/step - loss: 0.0176 - mean_absolute_error: 0.0884 - val_loss: 0.1331 - val_mean_absolute_error: 0.2502\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 208ms/step - loss: 0.0189 - mean_absolute_error: 0.1020 - val_loss: 0.1548 - val_mean_absolute_error: 0.2323\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 205ms/step - loss: 0.0146 - mean_absolute_error: 0.0891 - val_loss: 0.1307 - val_mean_absolute_error: 0.2061\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 190ms/step - loss: 0.0109 - mean_absolute_error: 0.0752 - val_loss: 0.1313 - val_mean_absolute_error: 0.2186\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 180ms/step - loss: 0.0106 - mean_absolute_error: 0.0760 - val_loss: 0.1445 - val_mean_absolute_error: 0.2025\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0075 - mean_absolute_error: 0.0622 - val_loss: 0.1358 - val_mean_absolute_error: 0.1891\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 170ms/step - loss: 0.0050 - mean_absolute_error: 0.0491 - val_loss: 0.1238 - val_mean_absolute_error: 0.1930\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 178ms/step - loss: 0.0056 - mean_absolute_error: 0.0579 - val_loss: 0.1288 - val_mean_absolute_error: 0.1799\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 169ms/step - loss: 0.0028 - mean_absolute_error: 0.0361 - val_loss: 0.1276 - val_mean_absolute_error: 0.1753\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 165ms/step - loss: 0.0026 - mean_absolute_error: 0.0361 - val_loss: 0.1207 - val_mean_absolute_error: 0.1742\n",
      "2/2 [==============================] - 0s 76ms/step - loss: 0.1207 - mean_absolute_error: 0.1742\n",
      "2308\n",
      "Done with seed: 39\n",
      "Keys: <KeysViewHDF5 ['DAS Data', 'Discharge', 'Times']>\n",
      "2308\n",
      "Epoch 1/1000\n",
      "7/7 [==============================] - 2s 175ms/step - loss: 0.8433 - mean_absolute_error: 0.7391 - val_loss: 0.4589 - val_mean_absolute_error: 0.5605\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 142ms/step - loss: 0.3931 - mean_absolute_error: 0.5018 - val_loss: 0.3695 - val_mean_absolute_error: 0.4520\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.3223 - mean_absolute_error: 0.4249 - val_loss: 0.3058 - val_mean_absolute_error: 0.4364\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.2495 - mean_absolute_error: 0.3858 - val_loss: 0.2853 - val_mean_absolute_error: 0.3796\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.2102 - mean_absolute_error: 0.3288 - val_loss: 0.2157 - val_mean_absolute_error: 0.3505\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 135ms/step - loss: 0.1548 - mean_absolute_error: 0.2892 - val_loss: 0.2039 - val_mean_absolute_error: 0.3356\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.1266 - mean_absolute_error: 0.2531 - val_loss: 0.1684 - val_mean_absolute_error: 0.3027\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0874 - mean_absolute_error: 0.2132 - val_loss: 0.1600 - val_mean_absolute_error: 0.2834\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.0683 - mean_absolute_error: 0.1799 - val_loss: 0.1503 - val_mean_absolute_error: 0.2628\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.0527 - mean_absolute_error: 0.1570 - val_loss: 0.1361 - val_mean_absolute_error: 0.2492\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 136ms/step - loss: 0.0398 - mean_absolute_error: 0.1356 - val_loss: 0.1351 - val_mean_absolute_error: 0.2488\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0351 - mean_absolute_error: 0.1267 - val_loss: 0.1344 - val_mean_absolute_error: 0.2483\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 138ms/step - loss: 0.0300 - mean_absolute_error: 0.1168 - val_loss: 0.1315 - val_mean_absolute_error: 0.2212\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0212 - mean_absolute_error: 0.0990 - val_loss: 0.1315 - val_mean_absolute_error: 0.2284\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0170 - mean_absolute_error: 0.0904 - val_loss: 0.1164 - val_mean_absolute_error: 0.2003\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0140 - mean_absolute_error: 0.0796 - val_loss: 0.1187 - val_mean_absolute_error: 0.2189\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0136 - mean_absolute_error: 0.0775 - val_loss: 0.1266 - val_mean_absolute_error: 0.2241\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0204 - mean_absolute_error: 0.0979 - val_loss: 0.1231 - val_mean_absolute_error: 0.2069\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.0215 - mean_absolute_error: 0.0975 - val_loss: 0.1343 - val_mean_absolute_error: 0.2392\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 137ms/step - loss: 0.0193 - mean_absolute_error: 0.0991 - val_loss: 0.1125 - val_mean_absolute_error: 0.1920\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0184 - mean_absolute_error: 0.0936 - val_loss: 0.1357 - val_mean_absolute_error: 0.2472\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 133ms/step - loss: 0.0290 - mean_absolute_error: 0.1229 - val_loss: 0.1271 - val_mean_absolute_error: 0.2282\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.0284 - mean_absolute_error: 0.1211 - val_loss: 0.1497 - val_mean_absolute_error: 0.2701\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0223 - mean_absolute_error: 0.1124 - val_loss: 0.1272 - val_mean_absolute_error: 0.2140\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - ETA: 0s - loss: 0.0113 - mean_absolute_error: 0.0763"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "IOPub message rate exceeded.\n",
      "The notebook server will temporarily stop sending output\n",
      "to the client in order to avoid crashing it.\n",
      "To change this limit, set the config variable\n",
      "`--NotebookApp.iopub_msg_rate_limit`.\n",
      "\n",
      "Current values:\n",
      "NotebookApp.iopub_msg_rate_limit=1000.0 (msgs/sec)\n",
      "NotebookApp.rate_limit_window=3.0 (secs)\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1000\n",
      "7/7 [==============================] - 2s 167ms/step - loss: 0.7284 - mean_absolute_error: 0.6984 - val_loss: 0.5834 - val_mean_absolute_error: 0.5707\n",
      "Epoch 2/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.3887 - mean_absolute_error: 0.4844 - val_loss: 0.4464 - val_mean_absolute_error: 0.5140\n",
      "Epoch 3/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.2794 - mean_absolute_error: 0.4056 - val_loss: 0.3174 - val_mean_absolute_error: 0.4064\n",
      "Epoch 4/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.1855 - mean_absolute_error: 0.3199 - val_loss: 0.2664 - val_mean_absolute_error: 0.3787\n",
      "Epoch 5/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.1445 - mean_absolute_error: 0.2674 - val_loss: 0.2465 - val_mean_absolute_error: 0.3533\n",
      "Epoch 6/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.1055 - mean_absolute_error: 0.2333 - val_loss: 0.2150 - val_mean_absolute_error: 0.3336\n",
      "Epoch 7/1000\n",
      "7/7 [==============================] - 1s 125ms/step - loss: 0.0885 - mean_absolute_error: 0.2150 - val_loss: 0.1576 - val_mean_absolute_error: 0.2822\n",
      "Epoch 8/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0762 - mean_absolute_error: 0.1848 - val_loss: 0.1768 - val_mean_absolute_error: 0.2826\n",
      "Epoch 9/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0576 - mean_absolute_error: 0.1712 - val_loss: 0.1571 - val_mean_absolute_error: 0.2575\n",
      "Epoch 10/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0437 - mean_absolute_error: 0.1462 - val_loss: 0.1265 - val_mean_absolute_error: 0.2384\n",
      "Epoch 11/1000\n",
      "7/7 [==============================] - 1s 125ms/step - loss: 0.0452 - mean_absolute_error: 0.1458 - val_loss: 0.1261 - val_mean_absolute_error: 0.2195\n",
      "Epoch 12/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 0.0477 - mean_absolute_error: 0.1454 - val_loss: 0.1651 - val_mean_absolute_error: 0.2815\n",
      "Epoch 13/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0570 - mean_absolute_error: 0.1660 - val_loss: 0.1245 - val_mean_absolute_error: 0.2331\n",
      "Epoch 14/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0432 - mean_absolute_error: 0.1406 - val_loss: 0.1150 - val_mean_absolute_error: 0.2042\n",
      "Epoch 15/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0369 - mean_absolute_error: 0.1418 - val_loss: 0.1338 - val_mean_absolute_error: 0.2312\n",
      "Epoch 16/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0248 - mean_absolute_error: 0.1103 - val_loss: 0.0971 - val_mean_absolute_error: 0.1864\n",
      "Epoch 17/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0270 - mean_absolute_error: 0.1057 - val_loss: 0.1033 - val_mean_absolute_error: 0.2086\n",
      "Epoch 18/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0267 - mean_absolute_error: 0.1133 - val_loss: 0.0974 - val_mean_absolute_error: 0.1817\n",
      "Epoch 19/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0146 - mean_absolute_error: 0.0884 - val_loss: 0.0947 - val_mean_absolute_error: 0.1819\n",
      "Epoch 20/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0179 - mean_absolute_error: 0.0918 - val_loss: 0.0928 - val_mean_absolute_error: 0.1840\n",
      "Epoch 21/1000\n",
      "7/7 [==============================] - 1s 125ms/step - loss: 0.0191 - mean_absolute_error: 0.0943 - val_loss: 0.1010 - val_mean_absolute_error: 0.1827\n",
      "Epoch 22/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0154 - mean_absolute_error: 0.0896 - val_loss: 0.1019 - val_mean_absolute_error: 0.1758\n",
      "Epoch 23/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0100 - mean_absolute_error: 0.0733 - val_loss: 0.0992 - val_mean_absolute_error: 0.1683\n",
      "Epoch 24/1000\n",
      "7/7 [==============================] - 1s 125ms/step - loss: 0.0122 - mean_absolute_error: 0.0792 - val_loss: 0.0914 - val_mean_absolute_error: 0.1751\n",
      "Epoch 25/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0134 - mean_absolute_error: 0.0827 - val_loss: 0.0921 - val_mean_absolute_error: 0.1737\n",
      "Epoch 26/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0138 - mean_absolute_error: 0.0850 - val_loss: 0.1008 - val_mean_absolute_error: 0.1756\n",
      "Epoch 27/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0108 - mean_absolute_error: 0.0755 - val_loss: 0.0987 - val_mean_absolute_error: 0.1671\n",
      "Epoch 28/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0135 - mean_absolute_error: 0.0878 - val_loss: 0.0970 - val_mean_absolute_error: 0.1991\n",
      "Epoch 29/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0144 - mean_absolute_error: 0.0894 - val_loss: 0.0928 - val_mean_absolute_error: 0.1577\n",
      "Epoch 30/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0104 - mean_absolute_error: 0.0814 - val_loss: 0.1037 - val_mean_absolute_error: 0.1842\n",
      "Epoch 31/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0083 - mean_absolute_error: 0.0670 - val_loss: 0.0891 - val_mean_absolute_error: 0.1522\n",
      "Epoch 32/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0115 - mean_absolute_error: 0.0766 - val_loss: 0.0887 - val_mean_absolute_error: 0.1737\n",
      "Epoch 33/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0118 - mean_absolute_error: 0.0795 - val_loss: 0.0947 - val_mean_absolute_error: 0.1513\n",
      "Epoch 34/1000\n",
      "7/7 [==============================] - 1s 135ms/step - loss: 0.0074 - mean_absolute_error: 0.0689 - val_loss: 0.0897 - val_mean_absolute_error: 0.1565\n",
      "Epoch 35/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0061 - mean_absolute_error: 0.0574 - val_loss: 0.0822 - val_mean_absolute_error: 0.1515\n",
      "Epoch 36/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0037 - mean_absolute_error: 0.0437 - val_loss: 0.0846 - val_mean_absolute_error: 0.1334\n",
      "Epoch 37/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0026 - mean_absolute_error: 0.0374 - val_loss: 0.0897 - val_mean_absolute_error: 0.1339\n",
      "Epoch 38/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0018 - mean_absolute_error: 0.0316 - val_loss: 0.0816 - val_mean_absolute_error: 0.1228\n",
      "Epoch 39/1000\n",
      "7/7 [==============================] - 1s 124ms/step - loss: 0.0020 - mean_absolute_error: 0.0328 - val_loss: 0.0825 - val_mean_absolute_error: 0.1345\n",
      "Epoch 40/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0016 - mean_absolute_error: 0.0281 - val_loss: 0.0835 - val_mean_absolute_error: 0.1191\n",
      "Epoch 41/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 9.6002e-04 - mean_absolute_error: 0.0233 - val_loss: 0.0851 - val_mean_absolute_error: 0.1248\n",
      "Epoch 42/1000\n",
      "7/7 [==============================] - 1s 134ms/step - loss: 0.0011 - mean_absolute_error: 0.0249 - val_loss: 0.0802 - val_mean_absolute_error: 0.1167\n",
      "Epoch 43/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0012 - mean_absolute_error: 0.0263 - val_loss: 0.0815 - val_mean_absolute_error: 0.1288\n",
      "Epoch 44/1000\n",
      "7/7 [==============================] - 1s 135ms/step - loss: 0.0013 - mean_absolute_error: 0.0254 - val_loss: 0.0820 - val_mean_absolute_error: 0.1173\n",
      "Epoch 45/1000\n",
      "7/7 [==============================] - 1s 132ms/step - loss: 9.9216e-04 - mean_absolute_error: 0.0238 - val_loss: 0.0849 - val_mean_absolute_error: 0.1280\n",
      "Epoch 46/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0013 - mean_absolute_error: 0.0278 - val_loss: 0.0820 - val_mean_absolute_error: 0.1214\n",
      "Epoch 47/1000\n",
      "7/7 [==============================] - 1s 124ms/step - loss: 0.0016 - mean_absolute_error: 0.0296 - val_loss: 0.0808 - val_mean_absolute_error: 0.1300\n",
      "Epoch 48/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0019 - mean_absolute_error: 0.0320 - val_loss: 0.0813 - val_mean_absolute_error: 0.1318\n",
      "Epoch 49/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0024 - mean_absolute_error: 0.0364 - val_loss: 0.0845 - val_mean_absolute_error: 0.1265\n",
      "Epoch 50/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0026 - mean_absolute_error: 0.0386 - val_loss: 0.0874 - val_mean_absolute_error: 0.1448\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 51/1000\n",
      "7/7 [==============================] - 1s 128ms/step - loss: 0.0036 - mean_absolute_error: 0.0448 - val_loss: 0.0800 - val_mean_absolute_error: 0.1242\n",
      "Epoch 52/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0036 - mean_absolute_error: 0.0463 - val_loss: 0.0841 - val_mean_absolute_error: 0.1554\n",
      "Epoch 53/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0054 - mean_absolute_error: 0.0570 - val_loss: 0.0838 - val_mean_absolute_error: 0.1255\n",
      "Epoch 54/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0048 - mean_absolute_error: 0.0564 - val_loss: 0.0923 - val_mean_absolute_error: 0.1603\n",
      "Epoch 55/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0056 - mean_absolute_error: 0.0575 - val_loss: 0.0795 - val_mean_absolute_error: 0.1258\n",
      "Epoch 56/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0052 - mean_absolute_error: 0.0581 - val_loss: 0.0872 - val_mean_absolute_error: 0.1583\n",
      "Epoch 57/1000\n",
      "7/7 [==============================] - 1s 125ms/step - loss: 0.0065 - mean_absolute_error: 0.0605 - val_loss: 0.0866 - val_mean_absolute_error: 0.1326\n",
      "Epoch 58/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0042 - mean_absolute_error: 0.0526 - val_loss: 0.0864 - val_mean_absolute_error: 0.1433\n",
      "Epoch 59/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0047 - mean_absolute_error: 0.0511 - val_loss: 0.0807 - val_mean_absolute_error: 0.1407\n",
      "Epoch 60/1000\n",
      "7/7 [==============================] - 1s 129ms/step - loss: 0.0034 - mean_absolute_error: 0.0466 - val_loss: 0.0830 - val_mean_absolute_error: 0.1328\n",
      "Epoch 61/1000\n",
      "7/7 [==============================] - 1s 125ms/step - loss: 0.0034 - mean_absolute_error: 0.0430 - val_loss: 0.0891 - val_mean_absolute_error: 0.1437\n",
      "Epoch 62/1000\n",
      "7/7 [==============================] - 1s 127ms/step - loss: 0.0023 - mean_absolute_error: 0.0390 - val_loss: 0.0799 - val_mean_absolute_error: 0.1176\n",
      "Epoch 63/1000\n",
      "7/7 [==============================] - 1s 131ms/step - loss: 0.0018 - mean_absolute_error: 0.0350 - val_loss: 0.0799 - val_mean_absolute_error: 0.1353\n",
      "Epoch 64/1000\n",
      "7/7 [==============================] - 1s 130ms/step - loss: 0.0017 - mean_absolute_error: 0.0333 - val_loss: 0.0835 - val_mean_absolute_error: 0.1195\n",
      "Epoch 65/1000\n",
      "7/7 [==============================] - 1s 126ms/step - loss: 0.0012 - mean_absolute_error: 0.0281 - val_loss: 0.0806 - val_mean_absolute_error: 0.1214\n",
      "2/2 [==============================] - 0s 60ms/step - loss: 0.0806 - mean_absolute_error: 0.1214\n",
      "2308\n",
      "Done with seed: 49\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import random\n",
    "import sys\n",
    "import eli5\n",
    "\n",
    "import d2d\n",
    "import importlib\n",
    "importlib.reload(d2d)\n",
    "import pandas as pd\n",
    "import os\n",
    "\n",
    "seeds = list(np.arange(0, 50, 1))\n",
    "residuals = {}\n",
    "\n",
    "for seed in seeds:\n",
    "    \n",
    "    SEED = seed\n",
    "    random.seed(SEED)\n",
    "    np.random.seed(SEED)\n",
    "    tf.random.set_seed(SEED)\n",
    "\n",
    "    file = \"/data/fast0/datasets/Rhone_data_continuous_highpass.h5\"\n",
    "\n",
    "    filt = 'Highpass'\n",
    "\n",
    "    history = {}\n",
    "    val_performance = {}\n",
    "    performance = {}\n",
    "\n",
    "    ww = 200\n",
    "    bs = 32\n",
    "\n",
    "    random.seed(SEED)\n",
    "    np.random.seed(SEED)\n",
    "    tf.random.set_seed(SEED)\n",
    "\n",
    "    input_columns = list(np.arange(0, 2308, 1))\n",
    "\n",
    "\n",
    "    linear_model, lstm_model, dnn_model, df_all_chan, das_data_all, f  = d2d.import_data(filename = file, input_columns = input_columns)\n",
    "\n",
    "    da = df_all_chan.to_numpy()\n",
    "\n",
    "    multi_step_window_shuffled = d2d.WindowGenerator(df_all_chan,\n",
    "                                                    input_width=ww,\n",
    "                                                    label_width=1, \n",
    "                                                    shift=0,\n",
    "                                                    label_columns=['Discharge'],\n",
    "                                                    input_columns=input_columns,\n",
    "                                                    shuffle=True,\n",
    "                                                    batch_size=bs)\n",
    "\n",
    "    history['lstm_whole_cable'] = d2d.compile_and_fit(lstm_model, multi_step_window_shuffled, learning_rate = 0.001)\n",
    "\n",
    "    val_performance['lstm_whole_cable'] = lstm_model.evaluate(multi_step_window_shuffled.val)\n",
    "    performance['lstm_whole_cable'] = lstm_model.evaluate(multi_step_window_shuffled.test, verbose=0)\n",
    "    \n",
    "    prediction_set = d2d.WindowGenerator(df_all_chan,\n",
    "                                    input_width=ww,\n",
    "                                    label_width=1, \n",
    "                                    shift=0,\n",
    "                                    label_columns=['Discharge'],\n",
    "                                    input_columns=input_columns,\n",
    "                                    shuffle=False,\n",
    "                                    batch_size=bs)\n",
    "    \n",
    "    predictions = lstm_model.predict(prediction_set.train).squeeze()\n",
    "    \n",
    "    predictions_real = (predictions * prediction_set.dis_std) + prediction_set.dis_mean\n",
    "    \n",
    "    real_dis = []\n",
    "\n",
    "    for i in prediction_set.train.as_numpy_iterator():\n",
    "        real_dis.append(i[-1])\n",
    "\n",
    "    real_dis = np.asarray(real_dis)\n",
    "    real_dis = real_dis.reshape(real_dis.shape[0] * real_dis.shape[1])\n",
    "\n",
    "    real_dis = (real_dis * prediction_set.dis_std) + prediction_set.dis_mean\n",
    "    \n",
    "    residuals['seed_'+str(seed)] = predictions_real - real_dis\n",
    "    \n",
    "    print('Done with seed: ' + str(seed))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "541d07c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "residuals_all = []\n",
    "\n",
    "for i in residuals.values():\n",
    "    residuals_all.append(i)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "02deed7b",
   "metadata": {},
   "outputs": [],
   "source": [
    "residuals_all = np.asarray(residuals_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "00dee678",
   "metadata": {},
   "outputs": [],
   "source": [
    "residuals_all = residuals_all.reshape(residuals_all.shape[0]*residuals_all.shape[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "9cb7b9da",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.9221535, -1.4198608, -1.1955094, ...,  1.3155489, -0.9258461,\n",
       "       -2.9804993], dtype=float32)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "residuals_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "f1328bdf",
   "metadata": {},
   "outputs": [],
   "source": [
    "rmse = np.sqrt(((residuals_all) ** 2).mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "dcc5cc07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.4937567"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9903fa25",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}