EXEMPLO_02 - MNIST Feed-Forward.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "3cabf146-204c-4abc-82d2-11cc591190c6",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "# Exemplo 1: MNIST Feed-forward\n",
    "\n",
    "**Nível de dificuldade:** <font color=green>básico</font>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1328c038-ad75-4e46-8068-cbbfc82c3ab4",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Objetivo\n",
    "\n",
    "Nesse exemplo, queremos criar um modelo capaz de <font color=\"red\">classificar</font> imagens de 28 por 28 píxels contendo números entre 0 e 9.\n",
    "Em outras palavras, queremos resolver um problema de classificação multiclasse.\n",
    "\n",
    "## Contextualização\n",
    "\n",
    "### MNIST\n",
    "\n",
    "O MNIST (**M**odified **N**ational **I**nstitute of **S**tandards and **T**echnology) é um banco de dados contendo um conjunto de 60.000 imagens de dígitos manuscritos.\n",
    "As imagens desse banco de dados são monocromáticas, ou seja, cada um dos seus pixels é representado por apenas um valor, que varia de 0 a 255. Além disso, estas\n",
    "imagens foram normalizadas, para caber em uma caixa delimitadora de 28x28 pixels.\n",
    "\n",
    "A imagem a seguir representa uma amostra dos dígitos presentes na base:\n",
    "\n",
    "![Exemplos MNIST](docs/_static/MnistExamples.png)\n",
    "\n",
    "\n",
    "O MNIST é comumente utilizado como exemplo introdutório para criação de modelos de deep learning.\n",
    "Alguns dos motivos pela escolha de usar a base do MNIST como exemplo são:\n",
    "\n",
    "- A base possui um conjunto considerável de imagens.\n",
    "- Devido ao tamanho pequeno e por serem imagens preto e branco, o banco de dados MNIST é relativamente pequeno e fácil de ser baixado.\n",
    "- Imagens já foram tratadas e normalizadas, sendo assim, não há necessidade de preprocessarmos as imagens.\n",
    "- O problema de classificação de dígitos manuscritos possui casos de uso reais, ao contrário de outros problemas introdutórios, que também são utilizados como exemplos introdutórios à problemas de aprendizado de máquina.\n",
    "\n",
    "\n",
    "### Redes Neurais\n",
    "\n",
    "\n",
    "Em ciência da computação, redes neurais artificiais (RNAs) são modelos computacionais inspirados nos mecanismos do sistema nervoso de animais.\n",
    "\n",
    "O diagrama a seguir representa alguns dos componentes que fazem parte uma rede neural:\n",
    "\n",
    "![Rede neural - diagrama simplificado](docs/_static/Artificial_neural_network.png)\n",
    "\n",
    "\n",
    "Cada nó, apresentado na imagem acima, representa a soma ponderada dos resultados dos nós anteriores.\n",
    "No caso da primeira camada, cada nó representa a multiplicação dos dados de entrada e um conjunto de híperparametros,\n",
    "chamdos de pesos (wheights), que serão ajustados ao longo do treino do nosso modelo.\n",
    "\n",
    "\n",
    "- $w\\,\\,\\rightarrow$ weight, ou peso.\n",
    "- $b\\,\\,\\rightarrow$ bias.\n",
    "\n",
    "\n",
    "Esses parametros são\n",
    "\n",
    "Durante a etapa de treino, estes parâmetros são ajustados, num processo chamado de **\"back propagation\"**.\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "Da mesma maneira que o cérebro humano opera através da transmissão de impulsos nervosos entre uma cadeia de neurônios, RNAs possuem um conjunto\n",
    "de funções comumente chamadas onde sinais são transmitidos\n",
    "\n",
    "#### Perceptron\n",
    "\n",
    "\n",
    "que são capazes de realizar o aprendizado de máquina bem como o reconhecimento de padrões. Redes neurais artificiais geralmente são apresentadas como sistemas de \"neurônios interconectados, que podem computar valores de entradas\", simulando o comportamento de redes neurais biológicas.\n",
    "\n",
    "#### Arquitetura Feed-forward\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5ae8dad-d83e-4bd9-9de8-47b4d256573b",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "# Importando as bibliotecas necessárias\n",
    "\n",
    "Para o exemplo que iremos resolver, utilizaremos algumas bibliotecas:\n",
    "\n",
    "- [Matplotlib](https://matplotlib.org/stable/index.html#installation)\n",
    "- [Numpy](https://numpy.org/install/)\n",
    "- [Tensorflow](https://www.tensorflow.org/install)\n",
    "\n",
    "```{important}\n",
    "Para rodar esse notebook, você precisa instalar os pacotes listados acima.\n",
    "```\n",
    "\n",
    "```{tip}\n",
    "Acesse os links listados acima, para maiores informações sobre as bibliotecas que serão utilizadas, incluindo instruções mais detalhadas de como instalá-las.\n",
    "```\n",
    "\n",
    "Para instalar os pacotes que iremos usar, execute o seguinte comando em uma janela da linha de comando:\n",
    "\n",
    "```bash\n",
    "$ pip install --user numpy tensorflow matplotlib\n",
    "```\n",
    "\n",
    ":::{attention}\n",
    "O código acima não irá instalar os pacotes, caso alguma versão deles já exista no seu environment. Dependendo de qual for essa versão instalada, isso pode vir a ser um problema. Para atualizar a versão dos pacotes, você pode incluir a opção <code>--upgrade</code> no comando apresentado anteriormente:\n",
    "\n",
    "```bash\n",
    "$ pip install --user --upgrade numpy tensorflow matplotlib\n",
    "```\n",
    ":::\n",
    "\n",
    "\n",
    "## Criando um environment para a demo\n",
    "\n",
    "As instruções abaixo mostram o passo-a-passo de como criar um environment novo usando o [Anaconda](https://anaconda.cloud/installers).\n",
    "\n",
    "<b>Passo 1:</b> Criando o novo environment\n",
    "\n",
    "Abra uma janela da linha de comando e, execute o seguinte comando:\n",
    "\n",
    "```bash\n",
    "$ conda create -N demo-pyimg -y python==3.9\n",
    "```\n",
    "\n",
    "O comando irá criar um novo environment chamado **demo-pyimg**, utilizando a versão 3.9 do Python.\n",
    "Você pode alterar o nome do environment, para o nome que você quiser. Quanto à versão do python,\n",
    "eu recomendo utilizar uma das seguintes versões: 3.8, 3.9, ou 3.10.\n",
    "\n",
    "<b>Passo 2:</b> Ativando o novo environment\n",
    "\n",
    "Para ativar o environment, execute o seguinte comando:\n",
    "\n",
    "```bash\n",
    "$ conda activate demo-pyimg\n",
    "```\n",
    "\n",
    ":::{note}\n",
    "Você também pode ver uma lista de environments disponíveis, utilizando o comando abaixo:\n",
    "\n",
    "```{code-block} bash\n",
    "---\n",
    "lineno-start: 1\n",
    "emphasize-lines: 1\n",
    "caption: List conda environments\n",
    "---\n",
    "$ conda env list\n",
    "```\n",
    "\n",
    "No meu caso, o comando acima retorna a seguinte lista:\n",
    "\n",
    "```{code-block} bash\n",
    "$ conda env list\n",
    "\n",
    "# conda environments:\n",
    "                         /Users/erikingwersen/opt/anaconda3\n",
    "base                     /opt/anaconda3\n",
    "DinoRunMl                /opt/anaconda3/envs/DinoRunMl\n",
    "FEM                      /opt/anaconda3/envs/FEM\n",
    "dataops                  /opt/anaconda3/envs/dataops\n",
    "demo                     /opt/anaconda3/envs/demo\n",
    "django-dashboard         /opt/anaconda3/envs/django-dashboard\n",
    "foobar                   /opt/anaconda3/envs/foobar\n",
    "ge_tutorials             /opt/anaconda3/envs/ge_tutorials\n",
    "gpt3-demos               /opt/anaconda3/envs/gpt3-demos\n",
    "nni                      /opt/anaconda3/envs/nni\n",
    "pandas-dev               /opt/anaconda3/envs/pandas-dev\n",
    "phoneix                  /opt/anaconda3/envs/phoneix\n",
    "py10                     /opt/anaconda3/envs/py10\n",
    "pyml                  *  /opt/anaconda3/envs/pyml\n",
    "sparky                   /opt/anaconda3/envs/sparky\n",
    "testml                   /opt/anaconda3/envs/testml\n",
    "tf-metal                 /opt/anaconda3/envs/tf-metal\n",
    "workspace                /opt/anaconda3/envs/workspace\n",
    "```\n",
    "\n",
    ":::"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "70596c2b-0934-4a6a-b452-a98893080591",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "pycharm": {
     "name": "#%%\n"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: numpy in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (1.22.3)\n",
      "Requirement already satisfied: tensorflow in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (2.8.0)\n",
      "Requirement already satisfied: matplotlib in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (3.5.1)\n",
      "Requirement already satisfied: absl-py>=0.4.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (1.0.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (1.1.0)\n",
      "Requirement already satisfied: wrapt>=1.11.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (1.13.3)\n",
      "Requirement already satisfied: protobuf>=3.9.2 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (3.19.4)\n",
      "Requirement already satisfied: tensorboard<2.9,>=2.8 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (2.8.0)\n",
      "Requirement already satisfied: flatbuffers>=1.12 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (2.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (4.1.1)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (0.2.0)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (3.3.0)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (0.24.0)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (1.6.3)\n",
      "Requirement already satisfied: gast>=0.2.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (0.5.3)\n",
      "Requirement already satisfied: keras<2.9,>=2.8.0rc0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (2.8.0)\n",
      "Requirement already satisfied: six>=1.12.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (1.16.0)\n",
      "Requirement already satisfied: setuptools in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (60.9.3)\n",
      "Requirement already satisfied: h5py>=2.9.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (3.6.0)\n",
      "Requirement already satisfied: keras-preprocessing>=1.1.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (1.1.2)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (1.44.0)\n",
      "Requirement already satisfied: tf-estimator-nightly==2.8.0.dev2021122109 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (2.8.0.dev2021122109)\n",
      "Requirement already satisfied: libclang>=9.0.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorflow) (13.0.0)\n",
      "Requirement already satisfied: cycler>=0.10 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from matplotlib) (0.11.0)\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from matplotlib) (2.4.2)\n",
      "Requirement already satisfied: packaging>=20.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from matplotlib) (21.3)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from matplotlib) (2.8.2)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from matplotlib) (1.3.2)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from matplotlib) (4.29.1)\n",
      "Requirement already satisfied: pillow>=6.2.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from matplotlib) (9.0.1)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from astunparse>=1.6.0->tensorflow) (0.37.1)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorboard<2.9,>=2.8->tensorflow) (1.35.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorboard<2.9,>=2.8->tensorflow) (0.6.1)\n",
      "Requirement already satisfied: werkzeug>=0.11.15 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorboard<2.9,>=2.8->tensorflow) (2.0.3)\n",
      "Requirement already satisfied: markdown>=2.6.8 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorboard<2.9,>=2.8->tensorflow) (3.3.6)\n",
      "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorboard<2.9,>=2.8->tensorflow) (0.4.6)\n",
      "Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorboard<2.9,>=2.8->tensorflow) (1.8.1)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from tensorboard<2.9,>=2.8->tensorflow) (2.25.1)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow) (4.8)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow) (0.2.8)\n",
      "Requirement already satisfied: cachetools<5.0,>=2.0.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow) (4.2.4)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.9,>=2.8->tensorflow) (1.3.1)\n",
      "Requirement already satisfied: importlib-metadata>=4.4 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from markdown>=2.6.8->tensorboard<2.9,>=2.8->tensorflow) (4.11.1)\n",
      "Requirement already satisfied: zipp>=0.5 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from importlib-metadata>=4.4->markdown>=2.6.8->tensorboard<2.9,>=2.8->tensorflow) (3.7.0)\n",
      "Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow) (0.4.8)\n",
      "Requirement already satisfied: idna<3,>=2.5 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow) (2.10)\n",
      "Requirement already satisfied: chardet<5,>=3.0.2 in /Users/erikingwersen/.local/lib/python3.8/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow) (4.0.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow) (2021.10.8)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow) (1.26.8)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.9,>=2.8->tensorflow) (3.2.0)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "!pip install numpy tensorflow matplotlib"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1db318c9-4850-4d04-b884-030224ad1a90",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Pacotes Opcionais\n",
    "\n",
    "Pacotes opcionais para execução do código da seção: [Arquitetura do modelo (Opcional)](#arq-modelo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "67dfa7d1-200d-4b06-a088-81f65fd50a47",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "!pip install graphviz pydot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "751d5e43-020d-4691-86ea-4d6ee18e5ab4",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Importando pacotes necessários"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5e237b80-d3fc-4352-9eec-b8ca4ec2f584",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from __future__ import annotations\n",
    "from typing import List, Any, Tuple, Iterable\n",
    "\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf13d2de-eeae-4223-abdd-440681fe352a",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "# Obtendo os dados do problema\n",
    "\n",
    "A próxima linha de código realiza o download dos dados a serem utilizados.\n",
    "A base será salva dentro do diretório `data/MNIST`.\n",
    "\n",
    ":::{note}\n",
    "As bases serão baixadas apenas uma vez. Execuções subsequentes carregarão o arquivo salvo em: `data/MNIST/mnist.npz`.\n",
    ":::\n",
    "\n",
    "**Fonte dos dados:** [MNIST homepage](http://yann.lecun.com/exdb/mnist/)\n",
    "\n",
    "**Função:** keras.datasets.mnist.load_data(path='mnist.npz')\n",
    "\n",
    "**Descrição:** Baixa e carrega o dataset MNIST.\n",
    "\n",
    "**Retorna:** tupla de tuplas de arrays [NumPy](https://numpy.org/doc/stable/user/quickstart.html), com os seguintes objetos: `(x_train, y_train), (x_test, y_test)`. Onde:\n",
    "\n",
    "\n",
    "- `x_train (np.ndarray[60000, 28, 28])`: 60.000 arrays de tamanho 28x28, representando o conjunto de imagens a serem utilizadas para treino do modelo.\n",
    "- `y_train (np.ndarray[60000])`: 60.000 labels, com valores entre 0 e 9, representando cada um dos dígitos de `x_train`.\n",
    "- `x_test (np.ndarray[10000, 28, 28])`: 10.000 arrays de tamanho 28x28, representando o conjunto de imagens a serem utilizadas para teste do modelo.\n",
    "- `y_test (np.ndarray[10000])`: 10.000 labels, com valores entre 0 e 9, representando cada um dos dígitos de `x_test`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6d0b5033-964f-4912-a57a-1eed7a28a748",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PosixPath('/Users/erikingwersen/Desktop/pyimg/data/MNIST')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_dir"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7c0fb63c-ef6c-4cf1-9827-2c58315869aa",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "data_dir = Path.cwd().joinpath(r'data/MNIST')\n",
    "data_dir.mkdir(parents=True, exist_ok=True)\n",
    "mnist_fpath = data_dir.joinpath('mnist').with_suffix('.npz')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9aa9ac7b-0a7b-4ac9-8668-19c694512e10",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# the data, split between train and test sets\n",
    "(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data(path=str(mnist_fpath))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5c874ff-1206-4b23-9039-a72014ad48e4",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Visualizando os dados\n",
    "\n",
    "Visualizando os dados do problema."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f7563c2e-0343-4f28-9d5f-d0ac4272c25a",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 8))\n",
    "plt.imshow(x_train[0], cmap='gray')\n",
    "plt.title(f'Digito \\\"{y_train[0]}\\\"', fontsize=20)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57d91afd-0472-42fe-9662-1d553711a771",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Refatorando...\n",
    "\n",
    "Criando a função `plot_imgs`, para ajudar na visualização das imagens."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "20d53d5e-6eba-433b-8117-4ebcfe66ee2d",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def plot_imgs(\n",
    "    imgs: List[np.ndarray],\n",
    "    labels: List[Any] | None = None,\n",
    "    labels_prefix: str = 'Label:',\n",
    "    cols: int = 5,\n",
    "    figsize: Tuple[int, int] | None = None,\n",
    "    maxsize: int = 30,\n",
    "    cmap: str = 'gray',\n",
    ") -> None:\n",
    "    \"\"\"Plot array of images.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    imgs: List[np.ndarray]\n",
    "        List of images.\n",
    "    labels: List[Any] | None, optional\n",
    "        List of labels for each image. If no label is specified,\n",
    "        then each figure is plotted without label.\n",
    "    cols: int, optional\n",
    "        Number of columns to plot. Defaults to 5.\n",
    "    figsize: Tuple[int, int], optional\n",
    "        Figure size. If no value specified, then function\n",
    "        uses `figsize = (12, 3 * rows)`.\n",
    "    maxsize: int, optional\n",
    "        Maximum number of images to plot.\n",
    "    cmap: str, optional\n",
    "        Colormap of each plot. Defaults to \"gray\".\n",
    "    \"\"\"\n",
    "    if imgs.ndim == 2:\n",
    "        imgs = [imgs]\n",
    "\n",
    "    num_imgs = len(imgs)\n",
    "    maxsize = min(num_imgs, maxsize)\n",
    "    rows = int(np.ceil(maxsize/cols))\n",
    "    cols = min(cols, num_imgs)\n",
    "\n",
    "    if figsize is None:\n",
    "        figsize = (12, 3 * rows)\n",
    "\n",
    "    fig, axes = plt.subplots(nrows=rows, ncols=cols, figsize=figsize)\n",
    "\n",
    "    current_col = 0\n",
    "    current_row = -1\n",
    "\n",
    "    for i, img in enumerate(imgs[:maxsize]):\n",
    "\n",
    "        _current_col = i // cols\n",
    "\n",
    "        if current_col != _current_col:\n",
    "            current_row = 0\n",
    "            current_col += 1\n",
    "        else:\n",
    "            current_row += 1\n",
    "        if rows <= 1:\n",
    "            ax = axes[current_row]\n",
    "        else:\n",
    "            ax = axes[current_col, current_row]\n",
    "\n",
    "        ax.imshow(img, cmap=cmap)\n",
    "        ax.axis('off')\n",
    "\n",
    "        if labels is not None and len(labels) >= i + 1:\n",
    "            curr_label = labels_prefix.rstrip(' ') + ' ' + str(labels[i])\n",
    "            ax.set_title(\n",
    "                curr_label,\n",
    "                fontsize=int((105*(figsize[0]/cols))/max(14, len(curr_label))),\n",
    "            )\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "563748da-a649-4f39-9e89-42b214778056",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1296 with 30 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_imgs(x_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3e531d9-470f-4e18-a5cb-50d16df864f0",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "# Normalizando os pixels de cada imagem"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dee00c90-b93c-4b1f-8212-021ae6e53765",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "Como as imagens já vem pré-processadas, o único tratamento que iremos aplicar é uma normalização dos píxels.\n",
    "Cada píxel possui um valor entre 0 e 255. Iremos converte-los em valores que variam de 0 a 1.\n",
    "Existem diversos métodos que podemos aplicar para normalização de dados.\n",
    "O método que iremos utilizar será o **min-max scaler**, que possui a seguinte fórmulação:\n",
    "\n",
    "$$X_{scaled}^{min-max} = \\frac{X - X_{min}}{X_{max} - X_{min}}$$\n",
    "\n",
    "Onde:\n",
    "\n",
    "$X\\,\\,\\rightarrow{}$ Cada píxel de cada imagem.\n",
    "\n",
    "$X_{min}\\,\\,\\rightarrow{}$ Píxel com o **menor** valor (0 no nosso caso).\n",
    "\n",
    "$X_{max}\\,\\,\\rightarrow{}$ Píxel com o **maior** valor (255 no nosso caso).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c05a2ae8-b30e-4df7-9687-17cbc4a0373a",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "max_pixel = x_train.max()  # 255\n",
    "\n",
    "# Convertendo cada pixel para um valor entre 0 e 1.\n",
    "x_train_norm, x_test_norm = x_train.astype(\"float32\") / max_pixel, x_test.astype(\"float32\") / max_pixel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d1f49f12-729b-4ae0-a08f-65a1b26765af",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: \u001B[94m\u001B[1m(60000, 784, 1)\u001B[0m\n",
      "Quantidade de imagens a serem utilizadas durante o treino: \u001B[94m\u001B[1m60,000 imagens\u001B[0m\n",
      "Quantidade de imagens a serem utilizadas durante o teste: \u001B[94m\u001B[1m10,000 imagens\u001B[0m\n"
     ]
    }
   ],
   "source": [
    "# Convertendo dimensão das imagens para: (784, 1)\n",
    "x_train_norm = x_train_norm.reshape(-1, 784, 1)\n",
    "x_test_norm = x_test_norm.reshape(-1, 784, 1)\n",
    "\n",
    "print(f\"x_train shape: \\033[94m\\033[1m{x_train_norm.shape}\\033[0m\")\n",
    "print(f\"Quantidade de imagens a serem utilizadas durante o treino: \\033[94m\\033[1m{x_train_norm.shape[0]:,} imagens\\033[0m\")\n",
    "print(f\"Quantidade de imagens a serem utilizadas durante o teste: \\033[94m\\033[1m{x_test_norm.shape[0]:,} imagens\\033[0m\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40107c21-87ae-44d2-aff3-1c94aab1157e",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "## Normalizando as labels\n",
    "\n",
    "Agora, precisamos converter cada um dos labels, em uma lista de binários.\n",
    "\n",
    "Por exemplo, o label \"5\", será convertido em:\n",
    "\n",
    "```\n",
    "\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "                     │\n",
    "                     └── sexta posição, logo, 5.\n",
    "```\n",
    "\n",
    "**Demais labels:**\n",
    "\n",
    "\n",
    "**Label 0:**\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[1., 0., 0., 0., 0., 0., 0., 0., 0., 0.]\n",
    "  │\n",
    "  └── primeira posição, logo, 0.\n",
    "```\n",
    "**Label 1:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "     │\n",
    "     └── segunda posição, logo, 1.\n",
    "```\n",
    "**Label 2:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "         │\n",
    "         └── terceira posição, logo, 2.\n",
    "```\n",
    "**Label 3:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "             │\n",
    "             └── quarta posição, logo, 3.\n",
    "```\n",
    "**Label 4:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "                 │\n",
    "                 └── quinta posição, logo, 4.\n",
    "```\n",
    "**Label 5:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "                     │\n",
    "                     └── sexta posição, logo, 5.\n",
    "```\n",
    "**Label 6:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "                         │\n",
    "                         └── sétima posição, logo, 6.\n",
    "```\n",
    "**Label 7:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "                             │\n",
    "                             └── oitava posição, logo, 7.\n",
    "```\n",
    "**Label 8:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "                                 │\n",
    "                                 └── nona posição, logo, 8.\n",
    "```\n",
    "**Label 9:**\n",
    "\n",
    "```\n",
    "  0   1   2   3   4   5   6   7   8   9\n",
    "[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]\n",
    "                                     │\n",
    "                                     └── décima posição, logo, 9.\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39af5a61-168c-48eb-ade6-342dc19dbfd9",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "### **Opção 1:** criando uma função própria para conversão"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c654f9ad-4cd4-4b1f-bf62-fc7aa5bce2ee",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def label_binarizer(labels: Iterable):\n",
    "    \"\"\"\n",
    "    Convert labels into lists of binary labels.\n",
    "\n",
    "    Function performs the following actions:\n",
    "\n",
    "    1. Retrieves set of unique labels, sorted.\n",
    "    2. Determines number of unique labels that exist.\n",
    "    3. Creates dictionary with each unique label's binarized translation.\n",
    "    4. Applies translation to every label.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    labels: Iterable\n",
    "        List of labels to binarize.\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    Tuple[np.ndarray, dict]\n",
    "        Numpy array with binarized labels as floats, and dictionary with translations.\n",
    "\n",
    "    Examples\n",
    "    --------\n",
    "    For example, consider the following list of labels:\n",
    "\n",
    "    .. code-block:: python\n",
    "\n",
    "        >>> example_labels = [2, 3, 2, 2, 5]\n",
    "        >>> binary_labels, translation_dict = label_binarizer(example_labels)\n",
    "        >>> print(binary_labels)\n",
    "        [[1. 0. 0.]\n",
    "         [0. 1. 0.]\n",
    "         [1. 0. 0.]\n",
    "         [1. 0. 0.]\n",
    "         [0. 0. 1.]]\n",
    "        >>> print(translation_dict)\n",
    "        {'2': [1, 0, 0], '3': [0, 1, 0], '5': [0, 0, 1]}\n",
    "\n",
    "    Function accepts multiple kinds of iterables as input:\n",
    "\n",
    "    .. code-block:: python\n",
    "\n",
    "        >>> import numpy as np\n",
    "        >>> example_labels = np.array([2, 3, 2, 2, 5])\n",
    "        >>> print(label_binarizer(example_labels)[0])\n",
    "        [[1. 0. 0.]\n",
    "         [0. 1. 0.]\n",
    "         [1. 0. 0.]\n",
    "         [1. 0. 0.]\n",
    "         [0. 0. 1.]]\n",
    "    \"\"\"\n",
    "\n",
    "    unique_labels = sorted(set(labels))\n",
    "    nunique_labels = len(unique_labels)\n",
    "\n",
    "    labels_binarized = {\n",
    "        label: [0]*nunique_labels\n",
    "        for label in unique_labels\n",
    "    }\n",
    "    for idx, unique_label in enumerate(unique_labels):\n",
    "        labels_binarized[unique_label][idx] = 1\n",
    "    return np.array([labels_binarized[label] for label in labels]).astype('float32'), labels_binarized\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c79f20b-4892-4f42-bb52-d593f59f0cd7",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### **Opção 2:** utilizando a função `keras.utils.to_categorical`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e7100181-8cf9-47ee-8787-e5a7f98c15cf",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Labels: 10\n"
     ]
    }
   ],
   "source": [
    "num_classes = len(set(y_train))      # 10 classes possíveis -> (0, 1, 2, 3, 4, 5, 6, 7, 8, 9).\n",
    "print(f'Labels: {num_classes}')\n",
    "\n",
    "y_train_norm = keras.utils.to_categorical(y_train, num_classes)\n",
    "y_test_norm = keras.utils.to_categorical(y_test, num_classes)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c46ea9c-553c-41ff-86d7-6a4a6c0e7ddb",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "# Criando o modelo\n",
    "\n",
    "Agora que temos o nosso input normalizado e, os labels convertidos em vetores unitários, podemos criar a rede neural que iremos utlizar para identificar o dígito de cada imagem.\n",
    "\n",
    "O modelo que iremos criar utilizará um conjunto de camadas, cada uma contendo metade da quantidade de nós da camada anterior."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d94c7629-4bee-4f04-b749-e5ec44026bf8",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 392)               307720    \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 196)               77028     \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 196)               0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 98)                19306     \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 49)                4851      \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 49)                0         \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 24)                1200      \n",
      "                                                                 \n",
      " Target_Layer (Dense)        (None, 10)                250       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 410,355\n",
      "Trainable params: 410,355\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "input_shape = tuple(x_train_norm.shape[1:])# -> Tamanho de cada imagem.\n",
    "input_dim = input_shape[0]\n",
    "\n",
    "model = keras.Sequential()\n",
    "model.add(keras.Input(shape=(input_dim,)))\n",
    "idx = 0\n",
    "\n",
    "while input_dim > 32:\n",
    "    input_dim = int(input_dim/2)\n",
    "    idx += 1\n",
    "    model.add(layers.Dense(input_dim, activation=\"relu\"))\n",
    "    if idx % 2 == 0:\n",
    "        model.add(layers.Dropout(.2))\n",
    "\n",
    "model.add(layers.Dense(num_classes, activation=\"softmax\", name='Target_Layer'))\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e33ea405-e2e6-43af-82c0-8520042e034b",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "<a id='arq-modelo'></a>\n",
    "## Arquitetura do modelo (Opcional)\n",
    "\n",
    "O código abaixo cria uma representação do modelo criado e salva a imagem gerada com o nome `MNIST_EXEMPLO_01.png`, dentro da pasta `outputs/MNIST`.\n",
    "\n",
    "**Observação:** a função `keras.utils.plot_model` requer duas bibliotecas adicionais: [graphviz](https://pypi.org/project/graphviz/) e [pydot](https://pypi.org/project/pydot/). Execute o commando abaixo para instalar as bibliotecas:\n",
    "\n",
    "```bash\n",
    "$ pip install graphviz pydot\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "28c31353-08c7-4fb4-a1de-8a20f8b07aaa",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: graphviz in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (0.19.1)\n",
      "Requirement already satisfied: pydot in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (1.4.2)\n",
      "Requirement already satisfied: pyparsing>=2.1.4 in /opt/anaconda3/envs/pyml/lib/python3.8/site-packages (from pydot) (2.4.2)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "!pip install graphviz pydot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a73c1173-13fc-4dab-adb6-173ae9d4d2f9",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output_dir = Path.cwd().joinpath('outputs/MNIST')\n",
    "output_dir.mkdir(parents=True, exist_ok=True)\n",
    "\n",
    "\n",
    "keras.utils.plot_model(\n",
    "    model,\n",
    "    to_file=output_dir.joinpath(\"MNIST_EXEMPLO_01\").with_suffix('.png'),\n",
    "    show_shapes=True,\n",
    "    show_dtype=True,\n",
    "    show_layer_names=True,\n",
    "    rankdir=\"TB\",\n",
    "    expand_nested=True,\n",
    "    dpi=96,\n",
    "    show_layer_activations=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5eb1e45f-cb97-4477-92a8-ee2e8348fe22",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Treinando o modelo\n",
    "\n",
    "Agora que criamos o modelo que iremos usar, podemos começar a etapa de treino, onde iremos passar o conjunto de imagens que separamos anteriormente.\n",
    "Durante essa etapa, os 410.355 parâmetros que compõe a nossa RNA serão ajustados gradativamente, durante a etapa de back propagation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ac2042e6-979c-46fa-8338-f2b5856e8b39",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "422/422 [==============================] - 4s 7ms/step - loss: 0.4129 - accuracy: 0.8761 - val_loss: 0.1212 - val_accuracy: 0.9655\n",
      "Epoch 2/20\n",
      "422/422 [==============================] - 3s 8ms/step - loss: 0.1416 - accuracy: 0.9609 - val_loss: 0.0972 - val_accuracy: 0.9703\n",
      "Epoch 3/20\n",
      "422/422 [==============================] - 4s 9ms/step - loss: 0.0940 - accuracy: 0.9735 - val_loss: 0.0872 - val_accuracy: 0.9770\n",
      "Epoch 4/20\n",
      "422/422 [==============================] - 4s 8ms/step - loss: 0.0701 - accuracy: 0.9794 - val_loss: 0.0777 - val_accuracy: 0.9787\n",
      "Epoch 5/20\n",
      "422/422 [==============================] - 4s 10ms/step - loss: 0.0562 - accuracy: 0.9839 - val_loss: 0.0977 - val_accuracy: 0.9737\n",
      "Epoch 6/20\n",
      "422/422 [==============================] - 5s 11ms/step - loss: 0.0483 - accuracy: 0.9862 - val_loss: 0.0762 - val_accuracy: 0.9803\n",
      "Epoch 7/20\n",
      "422/422 [==============================] - 5s 12ms/step - loss: 0.0395 - accuracy: 0.9878 - val_loss: 0.0845 - val_accuracy: 0.9788\n",
      "Epoch 8/20\n",
      "422/422 [==============================] - 5s 13ms/step - loss: 0.0372 - accuracy: 0.9888 - val_loss: 0.0707 - val_accuracy: 0.9825\n",
      "Epoch 9/20\n",
      "422/422 [==============================] - 6s 15ms/step - loss: 0.0305 - accuracy: 0.9911 - val_loss: 0.0943 - val_accuracy: 0.9800\n",
      "Epoch 10/20\n",
      "422/422 [==============================] - 6s 14ms/step - loss: 0.0276 - accuracy: 0.9921 - val_loss: 0.0941 - val_accuracy: 0.9790\n",
      "Epoch 11/20\n",
      "422/422 [==============================] - 7s 16ms/step - loss: 0.0256 - accuracy: 0.9921 - val_loss: 0.0804 - val_accuracy: 0.9818\n",
      "Epoch 12/20\n",
      "422/422 [==============================] - 7s 17ms/step - loss: 0.0207 - accuracy: 0.9937 - val_loss: 0.0803 - val_accuracy: 0.9825\n",
      "Epoch 13/20\n",
      "422/422 [==============================] - 7s 17ms/step - loss: 0.0226 - accuracy: 0.9935 - val_loss: 0.0712 - val_accuracy: 0.9842\n",
      "Epoch 14/20\n",
      "422/422 [==============================] - 7s 16ms/step - loss: 0.0175 - accuracy: 0.9948 - val_loss: 0.0917 - val_accuracy: 0.9810\n",
      "Epoch 15/20\n",
      "422/422 [==============================] - 7s 16ms/step - loss: 0.0192 - accuracy: 0.9941 - val_loss: 0.0967 - val_accuracy: 0.9808\n",
      "Epoch 16/20\n",
      "422/422 [==============================] - 7s 16ms/step - loss: 0.0178 - accuracy: 0.9947 - val_loss: 0.0761 - val_accuracy: 0.9843\n",
      "Epoch 17/20\n",
      "422/422 [==============================] - 6s 15ms/step - loss: 0.0166 - accuracy: 0.9952 - val_loss: 0.1014 - val_accuracy: 0.9807\n",
      "Epoch 18/20\n",
      "422/422 [==============================] - 5s 12ms/step - loss: 0.0133 - accuracy: 0.9960 - val_loss: 0.1033 - val_accuracy: 0.9807\n",
      "Epoch 19/20\n",
      "422/422 [==============================] - 6s 13ms/step - loss: 0.0116 - accuracy: 0.9971 - val_loss: 0.1031 - val_accuracy: 0.9815\n",
      "Epoch 20/20\n",
      "422/422 [==============================] - 6s 13ms/step - loss: 0.0147 - accuracy: 0.9959 - val_loss: 0.0953 - val_accuracy: 0.9827\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7fe38175fe80>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "batch_size = 128\n",
    "epochs = 20\n",
    "\n",
    "model.compile(loss=\"categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"])\n",
    "train_hist = model.fit(x_train_norm, y_train_norm, batch_size=batch_size, epochs=epochs, validation_split=0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d27dea1c-9b91-41c7-b4bb-3dd0849f1369",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "### Salvando métricas de treino com [Tensorboard](https://www.tensorflow.org/tensorboard/get_started)\n",
    "\n",
    "Você pode registrar a evolução do modelo durante o treino, utilizando o plugin [Tensorboard](https://www.tensorflow.org/tensorboard/get_started).\n",
    "Para utilizar o [Tensorboard](https://www.tensorflow.org/tensorboard/get_started), execute o comando:\n",
    "\n",
    "```bash\n",
    "$ pip install tensorboard\n",
    "```"
   ]
  },
  {
   "cell_type": "raw",
   "id": "b375ab3f-f4d2-4256-98c4-7fe4642721eb",
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    },
    "pycharm": {
     "name": "#%% raw\n"
    },
    "tags": []
   },
   "source": [
    "!pip install tensorboard"
   ]
  },
  {
   "cell_type": "raw",
   "id": "bb9c2462-d185-40a6-a8e8-44c4482a5ec8",
   "metadata": {
    "pycharm": {
     "name": "#%% raw\n"
    }
   },
   "source": [
    "%load_ext tensorboard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "586a383f-40bd-4068-a4d6-a5f0f92dc3e1",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import datetime\n",
    "import tensorflow as tf\n",
    "import io\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "73d54be0-6f35-4980-8adb-29d52f6f9bc2",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "log_dir = Path(f\"./logs/mnist-fit/{datetime.datetime.now().strftime('%Y%m%d-%H%M%S')}\")\n",
    "log_dir.mkdir(parents=True, exist_ok=True)"
   ]
  },
  {
   "cell_type": "raw",
   "id": "e471cac1-a564-4e27-b5a0-e3897de8817f",
   "metadata": {
    "pycharm": {
     "name": "#%% raw\n"
    }
   },
   "source": [
    "log_dir.unlink(missing_ok=True)\n",
    "log_dir.mkdir(parents=True, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "094d7c73-6a2c-4103-8720-0d44d2c15f37",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def plot_to_image(figure):\n",
    "    \"\"\"Converts the matplotlib plot specified by 'figure' to a PNG image and\n",
    "    returns it. The supplied figure is closed and inaccessible after this call.\"\"\"\n",
    "    # Save the plot to a PNG in memory.\n",
    "    buf = io.BytesIO()\n",
    "    plt.savefig(buf, format='png')\n",
    "    # Closing the figure prevents it from being displayed directly inside the notebook.\n",
    "    plt.close(figure)\n",
    "    buf.seek(0)\n",
    "\n",
    "    # Convert PNG buffer to TF image\n",
    "    image = tf.image.decode_png(buf.getvalue(), channels=4)\n",
    "\n",
    "    # Add the batch dimension\n",
    "    image = tf.expand_dims(image, 0)\n",
    "    return image\n",
    "\n",
    "\n",
    "def log_confusion_matrix(epoch, logs):\n",
    "\n",
    "    # Use the model to predict the values from the validation dataset.\n",
    "    test_pred_raw = model.predict(x_test_norm)\n",
    "    test_pred = np.argmax(test_pred_raw, axis=1)\n",
    "    test_labels = np.argmax(y_test_norm, axis=1)\n",
    "\n",
    "    # Calculate the confusion matrix.\n",
    "    cm = confusion_matrix(test_labels, test_pred)\n",
    "\n",
    "    # Log the confusion matrix as an image summary.\n",
    "    figure = ConfusionMatrixDisplay.from_predictions(test_labels, test_pred).plot()\n",
    "\n",
    "    # figure = plot_confusion_matrix(cm, labels=list(range(0, 9)))\n",
    "    cm_image = plot_to_image(figure.figure_)\n",
    "\n",
    "    cm_dir = log_dir.joinpath('cm')\n",
    "    cm_dir.mkdir(parents=True, exist_ok=True)\n",
    "    file_writer_cm = tf.summary.create_file_writer(str(cm_dir))\n",
    "\n",
    "    # Log the confusion matrix as an image summary.\n",
    "    with file_writer_cm.as_default():\n",
    "        tf.summary.image(\"Confusion Matrix\", cm_image, step=epoch)\n",
    "\n",
    "# Define the per-epoch callback.\n",
    "cm_callback = keras.callbacks.LambdaCallback(on_epoch_end=log_confusion_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "43e1127e-53d0-474c-bb01-de3a0fdb6e46",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "tensorboard_callback = keras.callbacks.TensorBoard(\n",
    "    log_dir=str(log_dir),\n",
    "    write_images=True,\n",
    "    histogram_freq=1,\n",
    "    update_freq='epoch',\n",
    "    profile_batch=(10, 20),\n",
    "    write_steps_per_second=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "5a7cae17-a7a1-4bc9-8313-536583a1c6d7",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "pycharm": {
     "name": "#%%\n"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "469/469 [==============================] - 3s 6ms/step - loss: 0.0101 - accuracy: 0.9971 - val_loss: 0.0120 - val_accuracy: 0.9973\n",
      "Epoch 2/15\n",
      "469/469 [==============================] - 3s 7ms/step - loss: 0.0107 - accuracy: 0.9969 - val_loss: 0.0024 - val_accuracy: 0.9993\n",
      "Epoch 3/15\n",
      "469/469 [==============================] - 3s 7ms/step - loss: 0.0087 - accuracy: 0.9974 - val_loss: 0.0032 - val_accuracy: 0.9990\n",
      "Epoch 4/15\n",
      "469/469 [==============================] - 3s 7ms/step - loss: 0.0087 - accuracy: 0.9975 - val_loss: 0.0044 - val_accuracy: 0.9987\n",
      "Epoch 5/15\n",
      "469/469 [==============================] - 4s 8ms/step - loss: 0.0088 - accuracy: 0.9973 - val_loss: 0.0026 - val_accuracy: 0.9993\n",
      "Epoch 6/15\n",
      "469/469 [==============================] - 4s 8ms/step - loss: 0.0089 - accuracy: 0.9975 - val_loss: 0.0019 - val_accuracy: 0.9993\n",
      "Epoch 7/15\n",
      "469/469 [==============================] - 4s 8ms/step - loss: 0.0079 - accuracy: 0.9979 - val_loss: 0.0017 - val_accuracy: 0.9997\n",
      "Epoch 8/15\n",
      "469/469 [==============================] - 5s 10ms/step - loss: 0.0074 - accuracy: 0.9980 - val_loss: 0.0012 - val_accuracy: 0.9997\n",
      "Epoch 9/15\n",
      "469/469 [==============================] - 4s 9ms/step - loss: 0.0094 - accuracy: 0.9973 - val_loss: 0.0041 - val_accuracy: 0.9988\n",
      "Epoch 10/15\n",
      "469/469 [==============================] - 5s 10ms/step - loss: 0.0095 - accuracy: 0.9971 - val_loss: 0.0014 - val_accuracy: 0.9997\n",
      "Epoch 11/15\n",
      "469/469 [==============================] - 6s 12ms/step - loss: 0.0076 - accuracy: 0.9979 - val_loss: 0.0026 - val_accuracy: 0.9992\n",
      "Epoch 12/15\n",
      "469/469 [==============================] - 6s 13ms/step - loss: 0.0081 - accuracy: 0.9978 - val_loss: 0.0010 - val_accuracy: 0.9997\n",
      "Epoch 13/15\n",
      "469/469 [==============================] - 6s 12ms/step - loss: 0.0063 - accuracy: 0.9983 - val_loss: 0.0038 - val_accuracy: 0.9990\n",
      "Epoch 14/15\n",
      "469/469 [==============================] - 5s 11ms/step - loss: 0.0059 - accuracy: 0.9982 - val_loss: 0.0046 - val_accuracy: 0.9990\n",
      "Epoch 15/15\n",
      "469/469 [==============================] - 6s 12ms/step - loss: 0.0093 - accuracy: 0.9970 - val_loss: 0.0011 - val_accuracy: 0.9998\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f89a961a3a0>"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAT4AAAEGCAYAAAD8EfnwAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAA+Z0lEQVR4nO2dd5xU5fX/32dmtvdC7x2xIRawBhWxoWgSEg0mxqgkRo2GxB7lG9MTNTG2/Aio2FBR7FIUNWiMKGChw9KWsgvsssv2NnN+f9y7sLTdOzv37s6wz/v1uq+de+e55569c+fMU89HVBWDwWDoSPja2wGDwWBoa0zgMxgMHQ4T+AwGQ4fDBD6DwdDhMIHPYDB0OALt7UBT0rID2qlHgut2i5a7b9NgiEVqqKROayUSG+efnaLFu4OOyi75pnaeql4QyfW8IKoCX6ceCfx+9tGu231qSB/XbRoMscgiXRCxjaLdQRbN6+mobFy39bkRX9ADoirwGQyGWEAJaqi9nYgIE/gMBkNYKBAithc+mMBnMBjCJoSp8RkMhg6EotSbpq7BYOhIKBA0TV1vWDEjjbWzUkFh8IQKjv5xOR/emkvZxjgA6sp9xKeFGP9GAbu+iefTe3MAUIUTbi6lz3nVjq81+aF8Ro4pp7QowE/PGeLa/3DS6DJ+9rvt+H3KnJnZvPxol6i169U9MHZjy1enxHofn6cTmEXkAhFZIyJ5InKn0/NK1saxdlYql8wqZPwbBWz5KImyzQHO/kcR498oYPwbBfQZW0Wf86oAyBpUzyWvWsfHTtvJp/flEGpw7uf8l7K5Z2K/sP+/5vD5lBv/uI3fTOzH9aOHcPb4UnoPqolau17cA2PXO5te2m0JBYKqjrZoxbPAJyJ+4DHgQmAYcKWIDHNybun6ODodV0cgSfEFoOvJtWyen7z3fVXYOCeZfuMqAfaWAwjWCoQ5PXP5olTKS9yt/A45oYrtm+IpzE+god7HR29kcur5e6LWrhf3wNj1zqaXdp0QcrhFK17W+E4B8lR1g6rWAS8C452cmDW4jh1LEqgp8dFQLWxdmERloX/v+zsWJ5CUEySj775q3a6v43nt4m68fmk3Tvtt8d5A2F7kdK1n1/b4vftFBXHkdquPWrsGg1MUJehwi1a8DA89gC1N9rcCIw8sJCKTgEkAud2tL3TmgAaOva6M+dd2JpCkZA+tQ5qE6A1vp9Dfru010un4Oi5/p4DS9QE+viOXHmdVEzAr1QwG11GF+uiNaY5o9yQFqjpVVU9S1ZPSsvfF4cETKrh0diEXPb+DhIwQ6X2tWk2oATa/l0y/i6oOaS9zQAOBZKV0bfwh328rigvj6NS9bu9+brd6igriotauweAcIehwi1a8DHzbgF5N9nvaxxxRXWy5VrHdz+b5yfS/xKrhbf80kYz+9aR03bdIunxLYO9gRsU2P3s2BEjtEcbohges+SqZHv3q6NKrlkBciNHjS/lsfkbU2jUYnKJASJ1t0YqXTd0vgEEi0g8r4F0B/MDpyR/e3ImaUh++AIyaspuEdOsubnw3hf4X79/M3bEkgWX/7mT16/mUU/9vN4nZzrtW73x8M8edWkFGdgPPLV7Jsw92Yd7MHMfnH4pQUHjsnh788YUN+Pww/8VsNq9NjMiml3a9uAfGbuz56pRors05QbwUGxKRi4B/AH7gSVX9Q3Pl+x+boiY7i8HgHYt0AWW6O6KodfRx8friO50dlT2u97YlqnpSJNfzAk/HPlX1XeBdL69hMBjaFgXqtd2HByIialduGAyG6EQRgu0/LhoRJvAZDIawCWls9/GZwGcwGMLCSlJgAp/BYOhQCMEY7+OLbe8NBkObY2Vg9jnaWkJEnhSRnSKyvMmxbBF5T0TW2X+z7OMiIv+0k558IyIjmpxztV1+nYhc3dJ1o6rGV7Q8wZOpJ/O2f+W6TYDzuw/3xK7BEM2oCnXqb7mgM54GHgWeaXLsTmCBqv7Zzup0J3AHVsKTQfY2EngCGCki2cAU4CSsuLxERN5U1ZLDXdTU+AwGQ9iEEEdbS6jqQmD3AYfHAzPs1zOAy5ocf0YtPgMyRaQbcD7wnqrutoPde0CzkpZRVeMzGAzRjzW44WmdqYuqFtivC4HGTLuHSnzSo5njh8UEPoPBECZhDW7kisjiJvtTVXWq05NVVUXE9eVlJvAZDIawaBzccEhRK5as7RCRbqpaYDdld9rHD5f4ZBsw+oDjHzV3AdPHZzAYwiao4mhrJW8CjSOzVwNvNDn+I3t0dxSwx24SzwPGikiWPQI81j52WEyNz2AwhIUi1Ks7oUNEZmLV1nJFZCvW6OyfgZdF5FpgM/A9u/i7wEVAHlAFXAOgqrtF5HdYGaEA7lfVAwdM9iPmAl9rFMa+d+zRZOY2MPXDNQAsfCuDZx/sypZ1ifzz3bUMPt5SZFvyn1Se/GN3GuqFQJxy/b3bGX5GBQC3fWcgu3cEiE+0uhv+9OL6Fq/rhQpWp+513PZwPpmdGkDh3edyeH16J1dsx4p6W6zdg0Z8PuWRuWspLojjvqv7u2LTS38Ph5uDG6p65WHeOvcQZRW48TB2ngSedHpdL8WGDpqYGCmtVRj7w/Mb9tvvO7SG+6Zt4thR++f1y8gOcv+MDfy/D9Zw28P5/PUXvfd7/47HNvPE+2t44v01ZOa2nOjUCxWsYIMw9f7uTBo9lFvGDeKSHxd1OPW2WLsHjVx2XRFb1kWeO7ERr/09HIqzZm4ETV3P8bKP72lamEsTLq1VGEvLCu6333tQLb0G1h5UbuCx1eR0tQJanyE11Nb4qKtt/YfnhQrW7p1x5C2zFOeqK/1syUt0RWwoltTbYu0eAOR2q+OUc8uY80K2K/bAW39bwq2VG+2FZ54dZmJiRLSlwtgn72Qw8Jhq4hP2jaQ/+Mve3DBmCM//vQvRIBnapWcdA46pZvXS5JYLt0CsqrfFyj342W+3M+333dCQe7Wg9vrMVCGoPkdbtNLufXxNVdYSifzhdYNNaxKZ/ofu/HHmvn68Ox7dTG63eqoqfPzuur68/0pWO3oIiclB7p22iX/d152qCteWD8UUsXIPRo4po7QoQN6yZI47taK93YkYa3Ajeu+3E9o98NmTGacCpEt2s/WotlAY27U9jvuv7cttD+fTve/+1wJITg1x9uWlrPmy/YK0P6DcO20TH8zO4r9zMl2xGWvqbbF0D4adXMmosWWcfO5K4hOU5LQgtz+ymb/eHNm69Pb8zGI9EWlMee+1wljFHj/3/qg/P7m7gKNP2TfwEWyAPcXWL1xDPSx6P52+Q73vRD40yuQHt7BlXSKzp7ozkgmxpt4WW/fgqT9146qThnH1yGH86YY+fP1JasRBz0t/W0IRQupsi1bavcYXDq1VGPvlJYPYszvAxBOH8cNfFZKWFeTx3/RgT3GAe3/YnwFHV/PHmRt486lctm+M5/mHuvL8Q10Ba9pKYnKIu38wgGCDEAzCiDMruHBiMY/c2avZ63qhgnX0KZWMmVDChpWJPP6eNT3nqT9144sP0iOyG0vqbbF2D7yiPf2N9RqfZyprTScmAjuAKao6vblz0iVbR8pB03cixqSlMhgs3FBZ63VMuk6eNcpR2cnD3utYKmvNTEw0GAwxjZjU8waDoWNhyUuaUV2DwdCBUBVCUTxHzwkm8BkMhrCJ5snJTjCBz2AwhIWVj8/08RkMhg5F7MtLdojA59W0k3s2fOW6zT/0H+66TYPHiEe1n2hYEH4IFKJ6crITOkTgMxgM7mHW6hoMhg5JNKeccoIJfAaDISystFSmqWswGDoYpo/PYDB0KKzsLKapazAYOhDWkjUT+NqcaFGrmnrBECbNtdIiVZf6ee3mvpRujSezZx2XP7qJpIwgResTePv23hSuSGL0rwoYdf2uvefXlPl5585e7FqbCALj/pLvma8t4YUaWiNu++ulyhp483zN+GwF1RV+QiFLLOnmi9y5x+2hsoap8R0eEekFPAN0wfqRmKqqD7thu1GtKjk12HJhBzSqVd11RX+KCuJ45N11fDYvg/wwFLE+/Vdn+p5Wzmk37OTTJzrzvyc6c86dBSRlBBl731bWvHdwgsj59/eg/7fK+M7jmwjWCfU1LT9Mbvh6KOa/lM2bT+Vy28NbIrJzIF7426iylrcsmaSUII/OXcvShWkR34NG3H6+Grl9wkDKXBRe8upZcEKsr9zwMmw3AL9S1WHAKOBGERkWqdFoVata+14Gx37H0lY69ju79wa6lNwGuh9fjf+A572mzEf+5ykM/551jj9eSUxv+YsWS2po4I2/XqmsgTfPl1e0l8pa46iukZc8BKpaoKpL7dflwCqgR6R2o1WtqrIojrTOljRlaqcGKoua1z4o3ZpAcnYDb9/em2njBvP2nb2oq2r544g1NTSv/XVTZQ28eb4AUOGPM9fz6Jw1XDixyBWT7fkshNTnaItW2sQzEekLnAAsOsR7k0RksYgsrudgrdumNFWrimZEQKT55UahBihckcyIiUVc9/Za4pNDfPqvzm3k4ZGB2yprXj5fky8fyE0XDOGeq/pz6Y+LOGZk7KqtGc0NB4hIKvAqcKuqlh34fjgqa9GsVpWSW0/5zgBpnRso3xkgOaeh2fLp3epJ71pPj+FVAAy9oNRR4Is1NTSv/PVCZc2r5wuguNCqme0pjuO/czIYOryK5YtSI7TZPs+CAg1RXJtzgqfei0gcVtB7XlVnR2ovmtWqBo8pY9mrVr/QslezGXxe830tqZ0aSO9WR/GGBAA2fZpGp0HN13jd8rUt8cZfb1TWvHq+EpKCJKUE974+8VvlbFoT+QBEez4Lsd7U9XJUV4DpwCpVfcir67hBa9Wqijck8s/ThnHWLYWc+rMdvHZTX756OYeMHnV8+9FNAFTsCvDk+MHUVvgRgc+f6sRP560mIS3E2P/bxuu39iFUL2T2rmPcX/P57N/N1/piSQ3NK3+9UlnziqxODUyZvhEAvx8+fD2TxR9F7mu7qaxFeTPWCV6qrJ0BfAwsA0L24btV9d3DneOVyppXmLRUBiCm0lK5obKWNbSznvPkdx2VnX36Ex1OZe0TiPHJPgaD4ZC4VeMTkV8C12F1HS4DrgG6AS8COcAS4IeqWiciCVhzg08EioHvq+qm1lw3ehvhBoMhKmlMRBrpqK6I9AB+AZykqscAfuAK4C/A31V1IFACXGufci1QYh//u12uVZjAZzAYwkIRGkI+R5sDAkCSiASAZKAAOAd4xX5/BnCZ/Xq8vY/9/rn2WELYmMBnMBjCJoQ42oDcxnm69jap0YaqbgMeAPKxAt4erKZtqao2zgfbyr6FDz2ALfa5DXb5Vo3AxWSSAoPB0I5oWH18RYcb3BCRLKxaXD+gFJgFXOCGiy1hAl8E/GHACa7bvHHdGtdtAjw2aLAndmNpRNPgDi6KDY0BNqrqLgARmQ2cDmSKSMCu1fUEttnltwG9gK120zgDa5AjbExT12AwhI1LS9bygVEikmz31Z0LrAQ+BBrny1wNvGG/ftPex37/A23lfDxT4zMYDGGhCEFnAxfN21FdJCKvAEuxsjl9ibV89R3gRRH5vX1sun3KdOBZEckDdmONALcKE/gMBkPYuJWPT1WnAFMOOLwBOOUQZWuACW5c1wQ+g8EQFhre4EZUYgKfwWAIGzWBz2AwdCxiP0lBzAU+r8RVvLCbkt7ALx/YQt8hNajCQ7/qzaolKY7P//rpTFa+nIEqHP29PRx/TSlFq+L56L4u1Ff5SO9Rz3kPFhKfFmLLJ8n874FcgvWCP0457Y5d9Dy1Oix/vbq3XgjteOGrlyJGR5bYkKnxHRYRSQQWAgn2dV6xOzJbjVfiKl7ZveH+bSz+MJ3fT+pHIC5EQlKo5ZNsitfGs/LlDL77aj7+OOWta3vQ55xKPrynK6fdsYseI6tZOSudL6dlMfKXxSRmBbn4/20jpUuQ4rXxvPWTnvz4kw2Or+e1cI2bQjte+eq1iNGRIjakCkG3U/O3MV7O46sFzlHV44HhwAUiMioSg16Jq3hhNzktyLEjK5k700pO2lDvo7LM+UNfsj6eLsfXEJek+ALQ/eRqNsxLpXRjHN1PsWpyvc6oYv08K4tvp6NrSeliJbvMHlRHQ40QrHX+cLaXcE1r8MpXL0WMvKA9P7MwlqxFJV6KDamqNgoLxNlbRNPxvRJX8cJu19617CkO8Ku/5/PYvDXc+rd8EpKcyxVmD6pj++Ikakp81FcLm/+TQkVhgOxBdWx832our5+TSkXhwanG189NpdPRNfgTnN9uT4VrXBbaaQuRHbdFjI4ksSHFauo62aIVT/v4RMSPteh4IPCYqh5SbAiYBJBIdIsIhYPfDwOPreKxe3uw5ssUfvbbrXz/pp0887dujs7PHljHiEm7efOangSSQuQeVYv44Jw/FfLx7zqz+LEc+p5bgS9u/+BWvC6e//0tl0uf2nYYy23P5MsHUlwYT0ZOPX9+cT1b8hIj1pvwErdFjCD27kHzxP7ghqdL1lQ1qKrDsdbbnSIixxyizFRVPUlVT4ojoVl7XomreGG3qCCOXQVxrPnSqp198k4mA48Nb7Bh2IQyvvd6Pt+euZWEjCCZ/erIGlDPpU9v43uv5zN4XDkZvff9wlcUBJjz8+6M+VshGX3C++X3UrjmUEI7kdnzzlcvRIzA/Xtg2Ww/4SlVZ1u00iZrdVW1FGv9XUSZF7wSV/HCbsmuOIq2x9NzQA0Aw88oJ39t84H9QKqKrdpG+fYAG+anMfiS8r3HNASLH8/h6CtKAagt8/H2pB6c+usiup1YE7a/Xt1bL4R2vBPZ8UbE6EgUGzJN3cMgIp2AelUtFZEk4DwiyJgK3omreGX3sXt7cMcjmwnEKYX58Tw4uXdY58+9qRs1JX58cXDWlB0kpIf4+ulMlj2fCcCAsRUc9V1LsXPZs5ns2RzHF49m88Wj1oDKpU9vIznHWb+iV/fAC6Edr3z1SsToSBMbskZ1Yzu/iZdiQ8dhZUv1Y9UsX1bV+5s7J9bEhrxIyXTjWpOWCojudtKBxNA9cENsKGlgd+33wKSWCwKrLv9thxMb+gZwP2GdwWBod6K5GeuEmFu5YTAY2hcluvvvnGACn8FgCJsY6og4JCbwGQyG8FDQGF+yZgKfwWAIG9PUNRgMHY5YGnQ/FIcNfCLyCM005VX1F554FEt48Ol7Ne3kR2u2eGL3mSG9PLEbU3gVBbyYJuOCq41rdWOZ5mp8i9vMC4PBEDsocKQGPlWd0XRfRJJVNfIFhgaDIeaJ9aZui+tORORUEVkJrLb3jxeRxz33zGAwRCmChpxt0YqTBXf/AM7HVixX1a+Bszz0yWAwRDvqcItSHI3qquoW2b+j1XlGTYPBcGShR/bgRiNbROQ0QEUkDrgFWOWtWwaDIaqJ4tqcE5wEvp8BDwM9gO3APOBGL506HJMfymfkmHJKiwL89Bx3VKpi0W6kylqrZqSyblYqqjBoQgXDfmwpBKx6NpU1z6cifuj5rWpOvH0PoXr49DfZ7F4ZhzYI/S+r5Niflrepv4fD51MembuW4oI47ru6vys2vfjMvHoOwDv1tpaJ7Rpfi318qlqkqhNVtYuqdlLVq1S12OkFRMQvIl+KyNuRuQrzX8rmnon9IjUT03YblbV+M7Ef148ewtnjS+k9yHni0ZK1cayblcpFs3ZwyRuFbP0oibLNAQo/S2DLgiQuebOQ8e8UMuxaK7htmptMqE649K0dXDx7B2tfSqViq/N07JH62xyXXVfEFpcVxbz4zLx6vhq5fcJAfj52aBsGPSDkcItSnIzq9heRt0Rkl4jsFJE3RCScn1fXmsbLF6VS7qI8XyzajVRZa8/6ALnH1RKw1du6nlxL/vwk1sxM5ZhJZfht7ZqkHOupFYGGaiHUAA01gi9OiUt13s7xSgkst1sdp5xbxpwXsiO21RQvPjOvnq92o3Een5MtSnEyqvsC8DLQDegOzAJmOjEuIj2Bi4FprXXQsD+RKmtlDq5nx5IEakp8NFQLWxcmUlnop2xTgJ2LE3h3QmfmXdWJom+sa/Q5v4pAkjLrjO7MPrsbR/+knIRM5z/lXimB/ey325n2+25RPWWiTfBAvc3RZWNcc8PJz1Cyqj7bZP85EbnNof1/ALcDaYcrcKSqrEUrmQMaOOa6ct6/thOBpBDZQ+vx+UCDQu0eHxe+vJPiZfEsvDWHyxcUUPRNPOJTJny8ndoyH/N+0Jlup9WQ1qv9BvZHjimjtChA3rJkjju1ouUTjmDaTb0tioOaEw5b4xORbBHJBuaIyJ0i0ldE+ojI7cC7LRkWkXHATlVd0ly5cFTWDO4oaw2aUMm42Tu44PldxGeESOvbQHKXBvqcV40I5B5XBz6oLfGx8e1kup9Zgy/Oav52GlFL8bL4li/ior8HMuzkSkaNLWPGopXc9cRmjj+jgtsf2RyRzVjFC/U2R7jU1BWRTBF5RURWi8gqe8FEtoi8JyLr7L9ZdlkRkX+KSJ6IfCMiI1rrfnNN3SVY63W/B/wUSyXtI+AG4PsObJ8OXCoim4AXgXNE5LnWOmqwcENZq7rY+tgrtvvJn59E/0sq6TWmmsJF1g9P2cYAoXpIyAqR0i1I4SJrAKG+Sij6OoGM/g1t6u+BPPWnblx10jCuHjmMP93Qh68/SeWvN/eJyGYs4pV6mxNEnW0OeBiYq6pDgeOxxgPuBBao6iBggb0PcCEwyN4mAU+01v/m1upGNAylqncBdwGIyGjg16p6VSQ273x8M8edWkFGdgPPLV7Jsw92Yd7MnEhMxpxdN5S1/nNzLrWlPnwBZeSUEuLTlYHfqeTTu7N5c1xXfHHK6X/ejQgMmVjBp3dl88bFXUFhwLcryRrqvI+uvZTAWosXn5lXz5dX6m0togIu9K2KSAbWKrAfA6hqHVAnIuOB0XaxGVgVrjuA8cAzaimkfWbXFrupakHY13aismYLgQ8D9j6xqvqM44vsC3zjmisXcyprMYRJSxWDeJCWalHo/YhV1hL69NJud93iqOzmG27bDDQddZmqqlMBRGQ4MBVYiVXbW4I1C2SbqmbaZQQoUdVMe0rcn1X1E/u9BcAdqhp2JqkWBzdEZApW9B2G1bd3IfAJ4DjwqepHWFHbYDAcCTgf3ChqRl4yAIwAblbVRSLyMPuatdZlVFXEYaM5DJxMZ/kucC5QqKrXYEXmtpFrNxgM0Yk7SQq2AltVdZG9/wpWINwhIt0A7L877fe3AU2bGD3tY2HjJPBVq2oIaBCRdNsJ074xGDoqLk1gVtVCrFwAjUtOzsVq9r4JXG0fuxp4w379JvAje3R3FLCnNf174Gwe32IRyQT+jdUGrwD+15qLGQyGIwMXG583A8+LSDywAbgGq0L2sohcC2zGmlkCVlfbRUAeUGWXbRUtBj5V/bn98l8iMhdIV9VvWntBg8FwBOBS4FPVr4BD9QEeNMppj+a6kiClObGhw04OFJERqrrUDQcMBkPs4f5wQ9vSXI3vwWbeU+Acl30xgDfKWng37eSoJd4svl91ovNJ0kcs0bzYNYoTEDihuQnMZ7elIwaDIUaI8rTyTjiCcuUYDIY2wwQ+g8HQ0ZAoTjLqBBP4DAZD+MR4jc9JBmYRkatE5D57v7eInOK9awaDIRpxmpklmkd+nazceBw4FbjS3i8HHvPMI4PBEP3EeOp5J03dkao6QkS+BFDVEnuWdZvTqXsdtz2cT2anBlB497kcXp/eyRXbXiiBeWGz54Aa7n5i0979rr3rePaBrrw2rXPEtiPxt/j5EKWvhxCBhIFCtyk+8n8eJGTnxQzuhsSjhV4P+Sl+JsSeOXYnURBqN8Lg9/34M5x/UeISQjw4O4+4eMUfUD5+J5NnH+gazr97SLz4zLx6br1Ub2uRKK7NOcFJ4KsXET/2vyoinXCon2QnIS3HEiBvaCZLgyOCDcLU+7uTtyyZpJQgj85dy9KFaeRHqLTVqAR21xX9KSqI45F31/HZvIyI7HphE2Dr+kR+Pnbo3ms8v2QF/52TGZHNSP2t36mUvBii/yw/vkRh6x1ByuYpfafve7y23hYk9VtWYMv5kY+cH1mNjfKFIXY/r2EFPYD6WuH2CQOoqfLjDygPvZ7HFx+ksXppSlh2muLVZ+bVczv/pWzefCqX2x72JuVYc0RzM9YJTpq6/wReAzqLyB+wUlL9MYxrnK2qwyMNegC7d8aRt8zS5aiu9LMlL9EV4RovlMC8UhdryvAzyinYnMDObZFXwCP1V4OgtaANitZAoEmFJlihVH6hpI0+OLiVzVXSz29Nk0ioqbJkLgNxij9OI57v69Vn5tVz227qbWqN6jrZohUna3WfF5ElWGvnBLhMVV2Ri4yELj3rGHBMNauXRi5QdCglsKEjItMu8MLmgYweX8pHr2e6YisSf+M6CzlX+Vh3cRBfAqSMElJP3febWv6RknKK4E/dP8CFqpWK/yld73Dy+3swPp/y6Ly1dO9bx1tP57Dmy9bX9qBtPjM3n9t25Uiv8YlIb6xMCG9hpYWptI85QYH5IrLEVlM7lP1JIrJYRBbXU+vIaGJykHunbeJf93WnqsK5uPWRRCAuxKixe1j4dmZ7u0KwTCn/jzLwLT+D5voJVcOed/f93JfNO3StruJjJfl4CbuZ20goJPz8vCFMPHEYQ4ZX0WdIdav/h7bgiHpu3cnH1244qSe/g/UvCFbq+X7AGuBoB+eeoarbRKQz8J6IrFbVhU0L2Gmop4KVer4lg/6Acu+0TXwwO8uVvi3wRgnMC5tNOfnscvKWJVNa5I7NSPytXKTE9YBAlhXA0s4Rqr5WMi6ChhKlZoWS+sDBv7F7DhMQw6WyzM/Xn6Zy8tnlbF6T1Go7Xn5mXjy37ckR38enqseq6nH230HAKTjMx6eq2+y/O7H6CSOc/6dMfnALW9YlMnuqO6O54I0SmBc2mzL6shLXmrkQmb9xXYXqZUqoWlFVqj5XEvpZAa18gZJ6huBL2D/ABcuVqqWH7vdzQkZ2AynplsJYfGKIEWdVsCUvssEC7z4zb55bQ+sJu2dUVZeKyMiWyolICuBT1XL79Vjg/lb4uJejT6lkzIQSNqxM5PH31gCW1OAXH0SmLOWFEpiX6mIJSUFGnFXOw3e4l3ElEn+TjhXSzxU2TgwiAUgYImR+2wpoZfND5Pz44N/X8g+V1FGCL6l1gS+7Sz2/fjgfnw98Plj4VgaL3o++5wC8e269Um9zRIzX+FpUWRORyU12fVg58XNU9fwWzuuPVcsDK8C+oKp/aO4co7KGZ2mpvEpxZNJSxRaLdEHEKmuJ3Xtp30mTWy4IrPnt5CVuzOhwGydPbVqT1w1YfX6vtnSSqm7AEiYyGAxHGjFe42s28NkTl9NU9ddt5I/BYIhyhNgf3Ggu9XxAVRtE5PS2dMhgMMQAR2rgAz7H6s/7SkTeBGYBlY1vqupsj30zGAzRSJRnXnGCkz6+RKAYS2OjcT6fAibwGQwdlShejuaE5gJfZ3tEdzn7Al4jMR7vDQZDJBzJNT4/kMr+Aa+RGP+3o5hoVtY6BF5NO7l85S5P7L42zEwgdoXYekwPornAV6CqEU04NhgMRyBRvg7XCc0FvuhNn2owGNqVI7mp28GXUBgMhsNypAY+Vd3dlo4YDIbYIZqTjDrByEsaDIbwOAL6+FqX+tZgMHRYJIzNkT0Rv4h8KSJv2/v9RGSRiOSJyEuN4mYikmDv59nv923t/xBTNT6v1Kq8suuFYpdXdr1S7Ir03uY9m8SmWYmg0HdCDQN/tC/L8rqnklj+t1Qu+m8RCVlKfbmw+I40qgr8aAMMuqaKPt92ltW7ES/urVeKcODdM9Yi7tb4bgFWAY15uv4C/F1VXxSRfwHXAk/Yf0tUdaCIXGGX+35rLuhpjU9EMkXkFRFZLSKrROTUSOw1qlVNGj2UW8YN4pIfF9F7UE3Efnpht1Gx6zcT+3H96CGcPb7UFV+9sjv/pWzumdgvYjsHEsm9LVvnZ9OsREa/VMI5r5VQ+FE8FZutR7aqwMfOT+NJ6hbcW37DC0mkDQhy7mslnDmjlGV/TSVUdzjrB+PVvW1UhLvhvCHccN4QThpdztARlS2f2E7+OsEtQXER6QlcDEyz9wVrldgrdpEZwGX26/H2Pvb759rlw8brpu7DwFxVHYqVoioikSKv1Kq8sOuVYpdXdr1S7Irk3pav95N9XD2BJPAFIPfkera/nwDAsr+kcsyvKvZPXyhKQ6WgCg1VQnyGImH8S94p47mvCAdto+R3WJxrbuQ2aurY24HaO/8AbmffIrgcoFRVG2fGbwV62K97AFsA7Pf32OXDxrPAJyIZwFnAdABVrVPVUrfse6VW5ZbdQyl2uRGkvbLbFoR7b9MGBSlaEkdtqdBQDYUL46ku8LN9QTxJnYNkDA3uV77/xBrKN/iZ861sFozP5ri7K5AwnnAv763Ppzz+3hpe+mYFXy5MjVgRDtrxWQhPXrJIVU9qsk1tNCMi44CdqrrEe6f3x8s+vn7ALuApETkeWALcoqr71fHtX4BJAIk4+0J4pVZ1RKlgRRmtubfpA4IMvq6aT6/LwJ+kZA5tIFQPa6cmc/q0g2s2Oz+JI2NoA2c8tYfKfB//vS6TnBNLiEtt/yHIRkW4lPQgU6ZvpM+Q6oiEkdodd27p6cClInIRVjKUdKxWYmZjWjygJ7DNLr8N6AVsFZEAkIGVQCVsvGzqBrDSWj2hqidgpbS688BCqjq18dcgjoQWjXqlVuW2Xa8Uu7xWb/OCSO5t3+/UcPYrpZz17B7i0pW0gUEqt/n54PIs5o3JpnqHjw+/k0XNLmHza4l0H1OHCKT2CZHcM0j5Buc/YG1xb5sqwkVKez4LbvTxqepdqtpTVfsCVwAfqOpE4EPgu3axq4E37Ndv2vvY73+gLWlnHAYvA99WYKuqLrL3X8EKhBHglVqV+3a9UuzyWr3NfSK7t7XFVide1XYf29+Pp/f4Gi7+pJjz39/N+e/vJqlLiLNfLSGxk5LcLcSuz6wvfk2RULHRT0qvYHPm98Ore+uFIhy087Pgra7uHcBkEcnD6sObbh+fDuTYxydziIqUUzxr6qpqoYhsEZEhqroGawncykhseqVW5YVdrxS7vLLrlWJXpPd20S0Z1JUKEgfH/6aC+PTDf5uG3FDF0rvTWDA+C1U4enIlCVnOv31e3VsvFOG89NcJbq/VVdWPgI/s1xs4hBStqtYAE9y4XosqaxEZFxmONUwdD2wArlHVksOVNyprhkZMWipvcENlLblTLx36bWcqa19OjV2VtVajql8BUfdPGwyG1nNEiw0ZDAbDYTGBz2AwdDQkxjKFH4gJfAaDITyOgOwsJvAZDIawMX18BoOhw2ESkXZkfB4saws5n3B7JOPVtJNf5K123eY/Bx3luk3PcKumZmp8BoOhQ+Ew5VQ0YwKfwWAIHxP4DAZDR8JMYDYYDB0SCcV25DOBz2AwhIeZx9e2eCUK5KYYzOQHNjNyzB5LtGfMMADOvLiEH04uoNegGn4xbgjrvoks+24siQ2BN/5GavOrp7NY/lImKBz9/VJOuKaEXasS+PDertRXCWk9Gjj/oe0kpIVY/UY6S6dl7z23aHUCV76xiU7DnAsZXX79Ti68cjeqsHF1Ig9O7k19beRZ4VLSG/jlA1voO6QGVXjoV71ZtSTy7M4tEevTWbxMPT9ERL5qspWJyK2R2PRKbMhNMZj5s7K556qB+x3btCaR+6/vz7JFqRH7GmtiQ174G6nN4rXxLH8pk+/P3sQP3t7Ipg9TKd0Ux4K7u3LabTuZ+O4mBowt3xvsho4v4wdvbeIHb21i7AMFpPesDyvo5XSt47KfFHHTRYP56blD8fth9PjDJikKixvu38biD9O57ltHccN5Q8hf13IyX1fwNh+f53gW+FR1jaoOV9XhwIlAFfBaJDa9EhtyUwxm+aI0ykv3n9+3JS+JrRvcyZMWa2JDXvgbqc3deQl0Pb6auCTFF4Aep1Sxfn4apRvj6XGKJV/Z+/RK8uamHXTu2rfSGDyuLGyf/QElITGEz68kJIUoLow8U3JyWpBjR1Yyd6YVoBvqfVSWtU0jzi2VtfairQTFzwXWq+pmtwy6LTbkhRiMF8Sa2JAX/kZqM2dwLdsXJ1Nd4qO+Wtj0USrlBXFkD6plw/tWrXzdnDQqCg8OImvfSWfIJeEFvuLCeF75V2ee/XwlM79cTmWZn6ULI09E2rV3LXuKA/zq7/k8Nm8Nt/4tn4SkNpgAr4Cqsy1KaavAdwUw81BviMikRum5epw1H7wQBWoUg5l44jCGDK+iz5Dqlk8yxCTZA+s4cVIxr/+4N2/8pBedhtXg8ytj/lzIN89lMXN8X+orffgPqJQVfpVIXFKInMFhiPUCqRkNnHr+Hq4eNYwfjDiGxOQg53x7d8T/h98PA4+t4u1ncrnx/CHUVPn4/k07I7brhDBU1qISzwOfiMQDlwKzDvV+tIgNNeKmGIwXxJrYkBf+umHz6O/t4co3NvHdmfkkpAfJ7FtH9oA6Lp+xhSvf2MTgS8rI6L1/gFv7djqDx4X/XJxwZgWF+fHs2R0g2CD8d04mw06KXFC8qCCOXQVxe1snn7yTycBjvf/BbpzHZ5q6zXMhsFRVd0RuyhuxIa/EYLwg1sSGvPDXDZtVxVZLoXx7gPXz0xhyadneYxqCLx7L5ZgrS/eW15DV/G1N/97ObXEcNaKKhMQQoAw/o5z8dZE/XyW74ijaHk/PAdbAzvAzyslf2waDG06buVHc1G2LntArOUwzN1y8EhtyUwzmzkc3ctyp5ZZozxfLePbBbpSXBvj577aQkd3A72asZ/2KJO65alCr7Mea2JAX/rph890be1Bd4scfp4z+vx0kpIf46uksvnkuC4ABY8sZ9t19AybbPk8mtWsDGb3D759c82UKH7+TwWPz1hBsEPJWJDHn+cjvLcBj9/bgjkc2E4hTCvPjeXByb1fstkQ01+ac4LXYUAqQD/RX1RaH3WJObMhkZ4k5Onp2lkWh9yMWG0rL7KknnHWLo7Ifv3V7hxQbqsTSxTQYDEcQsV7ji6mVGwaDIQpQIBjbkc8EPoPBEDamxmcwGDoeUTxi6wQT+AwGQ9iYGp/BYOhYRHkCAieYwBcJZupJzPHPgUNdtzlx9RbXbQI8P7SnJ3YjRQAxgxsGg6GjITHex9dWSQoMBsORgtNcfC3ERhHpJSIfishKEVkhIrfYx7NF5D0RWWf/zbKPi4j8U0TyROQbERnR2n/BBD6DwRAmrq3VbQB+parDgFHAjSIyDLgTWKCqg4AF9j5Y6/4H2dsk4InW/gcm8BkMhrBxIzuLqhao6lL7dTmwCugBjAdm2MVmAJfZr8cDz6jFZ0CmiHRrjf+mj89gMISP8z6+XBFZ3GR/qqpOPbCQiPQFTgAWAV1UtcB+qxBoFFTpATQdSdpqHysgTEzgMxgM4aFhjeoWtZSkQERSgVeBW1W1TGRfDgVVVRH3Zw3GVOCLNSUwN9XbDsTnUx6Zu5bigjjuu7q/KzZjSb3NC18hcn9XP5NK3qwUUBg4oZKhV1cAsObZFNa+kIr4ofu3ahhx2x6Kvonj8/usNFiqcNxNZfQ6z7lokpffhxZxKRSJSBxW0HteVWfbh3eISDdVLbCbso1ppbcBvZqc3tM+Fjae9vGJyC/t0ZrlIjJTRCJKxBZLSmDgrnrbgVx2XRFbXEhm2Ugsqbd55StE5m/p2gB5s1K44OWdXPT6DrZ9lEj5Zj+FnyWw9YMkLnpjB+Pe3sGwn1hZnDMHNXDBKzu56PWdnPPvIhZNySLU0Da+RoqoOtqatWFV7aYDq1T1oSZvvQlcbb++GnijyfEf2aO7o4A9TZrEYeGlvGQP4BfASap6DODH0t5oNbGkBGbhnnpbU3K71XHKuWXMeSG75cIOiSX1Nu8+r8j83bMhjtzj6gjY6m2dT64l/70k1r2YwrDry/Hb+kiJOZYYRWM5gGCdIGFmyfPq++AId0Z1Twd+CJzTRIb2IuDPwHkisg4YY+8DvAtsAPKAfwM/b637Xt+1AJAkIvVAMrDd4+u1ikOpdg0dUeWKbZ9PeXTeWrr3reOtp3NcUW/72W+3M+333UhOdU/Nxct74DbR6mvmoHq+/ns6tSU+/InK9v8kkn1MPeWbAuxanMDX/8jAH6+MuKOUnGOtTM5FX8fz2T1ZVG73c9pfdu8NhFGNAi48eqr6CdZCkENxUEZitbIm3xj5lb3V1d0GPICVgbkAq1o6/8ByrVFZiyXcVm8bOaaM0qLAXn1hQ/SQMaCBYdeX88G1uXxwfS5ZR9UjfiUUFGr3+Dj/pZ2ccHspH9+as7cylHt8HePe3sEFs3ayYmo6wRj4CgjOmrnRvLrDy6ZuFta8m35AdyBFRK46sFy4Kmte0BbKZW6ptw07uZJRY8uYsWgldz2xmePPqOD2RyKXK44l9bZo9nXgd6u4cPZOxj63i/j0EOl9G0juEqTXedWIQO5x9YgPakv2/+plDGggkByidG10/B8tEgo526IULwc3xgAbVXWXqtYDs4HTPLxeq/FKucwL9ban/tSNq04axtUjh/GnG/rw9Sep/PXmPhH7GkvqbdHsa02x9ZWq3O5ny3tJ9B1XRc8x1ez43PpRL9sYIFQPCVkhKrb69w5mVGzzU7YhjpSeMZD4orGp62SLUrzsUcgHRolIMlCN1WZf3PwpzRNLSmDgrnqb18SSeptXvrrh78Jf5FBb6sMXUE6+r5T4dGXAtyv57J4s3r6kC7445dQ/lyACO5cksPLfafgCCj44eUoJiVnOo4VX3wcnRHMz1gleq6z9Fvg+1pq8L4HrVPWwvRgxp7JmMAATV2/1xK4XaakW6YKIVdYykrvrqYOudVR23je/75Aqa1OAKV5ew2AwtDXRLRbuhFgYPDcYDNGEUVkzGAwdkVjv4zOBz2AwhI8JfAaDoUOhQMgEPoPB0KEwgxsGtwl3pbpjux7NVTdKc56pod2wLs91mxsvcyeLjQl8BoOhY6FAMIqXZTjABD6DwRAmCmoCn8Fg6GiYpq7BYOhQmFFdg8HQITE1PoPB0OEwga/t8EpVqlP3Om57OJ/MTg2g8O5zObw+vVPEdr1SApvx2QqqK/yEQhBsEG6+yJ17cdm1O7nwyiJEYM4Lubw2vXPENmNJZc2r5yBStb1vns5g5cvpoHDU98o4/po9FK2M5z/3dSZYK/gCypn/t4sux1uJj7YtSuK/v88l1ACJWSEue6FVQmSHRxWCsT2NydPAJyK3ANdj5dX/t6r+IxJ781/K5s2ncrnt4S0tFw6DYIMw9f7u5C1LJiklyKNz17J0YRr5EaiYNSqB3XVFf4oK4njk3XV8Ni8jIptNuX3CQMpcFJrpM6SaC68s4hfjhlJfL/zxuTwWLUhn+6bI/PXiM/Pq3nrxHMA+tb2aKj/+gPLQ63l88UEaq5e2rL9SvDaelS+n851Xt+KPU96+tjt9z6nif3/N5aSbd9PnW1Vs/iiZz/6ay/jnt1Fb5uPjKZ24+MntpHVvoKrYH5HvhyXGa3xepp4/BivonQIcD4wTkYGR2PRKVWr3zri9GhbVlX625CWS260+IpteKoF5Qe+BNaz+KoXaGh+hoPDNZ6mcfmFpxHZjSWXNi+fAovVqe6Xr4+hyfC1xtiJb95Or2TAvBRGor7C+vnXlPpI7W6mc172VSr+xFaR1t/aTczyqmbmjstZueJl6/ihgkapWqWoD8B/g2x5ezxW69KxjwDHVrF4amZjPoZTA3PkSASr8ceZ6Hp2zhgsnFrlictOaRI45pYK0zAYSEkOcfE4Znbq75K/LeHpvbdx6Dhrx+ZTH31vDS9+s4MuFqY7V9rIH1VGwOJGaEh/11UL+f1KoKAxw+j27+N9fcnjmzD787y+5jPp1MQClG+OpLfPxxsQezLqsJ2teS3PF//1Ra1TXyRaleNnUXQ78QURysFLPX8QhUs+LyCRgEkAi7asclpgc5N5pm/jXfd2pqvCoieACky8fSHFhPBk59fz5xfVsyUtk+aLUiGxuyUvi5ce78KcX1lFT5WfDiqQOuxrNi+egUW0vJT3IlOkb6TOkms1rklo8L2tgPSdMKuGta7oTl6TkHFWL+GDFCxmcdncRAy6oJO/dVD68uzOXztiOBmHX8kQufWYbDTXCa9/rSZfhNWT2c/GHQUFjfAKzl/KSq4C/APOBucBXwEFfpWhQWQPwB5R7p23ig9lZ/HdOZsT2vFQCKy60ajt7iuP475wMhg53R1N23ou53HTRUfz6u4Op2ONn6wZ3+iPdxst76/ZzcCCtUds7akI5E17fymUzt5GQESSzXx1rXkuj//mVAAy4sIKdX1ufVUrXBnqdWUVcspKUHaLbyTUUr45vznzrCIacbVGKl01dVHW6qp6oqmcBJcBaL6/XepTJD25hy7pEZk+NfBQPvFMCS0gKkpQS3Pv6xG+Vs2mNOwEqI8eqFXTqXsfpF5by4etZrth1G+9U1tx/DiBytb3GAYry7QE2zk9l0CUVJHcOsv1zq8a47X9JZPS1fgj6nVtJ4ZJEQg1QXy3s+DqBzAEud1moxry8pNejup1VdaeI9Mbq3xsViT2vVKWOPqWSMRNK2LAykcffWwNYMo5ffNB6RTSvlMCyOjUwZfpGAPx++PD1TBZ/5I5y231TN5CWFSTYIDx6Ty8qyyJ/PGJJZc2L5wAiV9ubd1NXakv8+OKUM6fsIiE9xOg/7OST3+eiQcEfr4z+/S7Aahr3OrOKl8f1Bp9y1IQycgbXtXCFVhDFAxdO8Fpl7WMgB6gHJqvqgubKG5U1TFoqw168SEt1+2VryFtWFZnKmj9XRyVd7Kjs/MpnOqTK2ple2jcYDO1BdE9VcUJMrdwwGAxRgElSYDAYOhoKaIwvWfN0VNdgMByBqJ2I1MnWAiJygYisEZE8EbmzDbwHTI3PYDC0AnWhqSsifuAx4DxgK/CFiLypqisjNt4CpsZnMBjCx50a3ylAnqpuUNU64EVgvOe+4/F0lnARkV3AZgdFcwF3Fqkau21h09j1zma4dvuoakSzs0Vkrn1NJyQCTaXdpqrqVNvOd4ELVPU6e/+HwEhVvSkS/5wQVU1dpx+IiCz2Ym6QsRtbvsaa3VjytTlU9YK2upZXmKauwWBoL7YBvZrs97SPeY4JfAaDob34AhgkIv1EJB64AnizLS4cVU3dMJhq7HpmN5Z8jTW7seSr56hqg4jcBMwD/MCTqrqiLa4dVYMbBoPB0BaYpq7BYOhwmMBnMBg6HDEX+LxY4iIiT4rIThFZ7oY922YvEflQRFaKyApbcc4Nu4ki8rmIfG3b/a0bdpvY94vIlyLytos2N4nIMhH5SkQOkh9opc1MEXlFRFaLyCoROdUFm0NsHxu3MhG51QV3EZFf2p/XchGZKSKuZI8VkVtsmyvc8rVDoKoxs2F1gK4H+gPxwNfAMBfsngWMAJa76Gs3YIT9Og0r+7QbvgqQar+OAxYBo1z0ezLwAvC2izY3AbkuPwszgOvs1/FApgfPWiHWhN9IbfUANgJJ9v7LwI9dsHsMlrZNMtZA5fvAQDfvw5G6xVqNz5MlLqq6ENgdqZ0DbBao6lL7dTmwCusLEKldVdUKezfO3lwZoRKRnsDFwDQ37HmFiGRg/VhNB1DVOlUtdfky5wLrVdXJSiInBIAkEQlgBartLtiMSSXDaCDWAl8PoKky9VZcCCZeIyJ9gROwamdu2POLyFfATuA9VXXFLvAP4HbAbbEEBeaLyBJbVS9S+gG7gKfsZvk0EXGm1+icK4CZbhhS1W3AA0A+UADsUdX5LpheDpwpIjkikoylZNirhXMMxF7gizlEJBV4FbhVVcvcsKmqQVUdjjXT/RRbvD0iRGQcsFNVl0Rq6xCcoaojgAuBG0XkrAjtBbC6Jp5Q1ROASsC1lEb2ZNpLgVku2cvCapn0A7oDKSJyVaR21aGSoeFgYi3wtdsSl9YgInFYQe95VZ3ttn27efch4MbaydOBS0VkE1YXwjki8pwLdhtrPKjqTuA1rC6LSNgKbG1S030FKxC6xYXAUlXd4ZK9McBGVd2lqvXAbOA0NwxrzCgZRhexFvjabYlLuIiIYPVBrVLVh1y020lEMu3XSVi5zFZHaldV71LVnqraF+u+fqCqEddKRCRFRNIaXwNjsZpokfhaCGwRkSH2oXMBN3O4XYlLzVybfGCUiCTbz8W5WH2+ESMine2/jUqGL7hh90gnppasqUdLXERkJjAayBWRrcAUVZ0eodnTgR8Cy+z+OIC7VfXdCO12A2bYSRx9wMuq6trUEw/oArxmfd8JAC+o6lwX7N4MPG//AG4ArnHBZmNwPg/4qRv2AFR1kYi8AiwFGoAvcW+Z2asi0qhkeKMHgzxHJGbJmsFg6HDEWlPXYDAYIsYEPoPB0OEwgc9gMHQ4TOAzGAwdDhP4DAZDh8MEvhhCRIJ21pDlIjLLXqbUWltP2ypX2Eu+hjVTdrSIhD3h1s7KcpAa1+GOH1Cmorn3D1H+/0Tk1+H6aOiYmMAXW1Sr6nBVPQaoA37W9E17AXzYqOp12ryI82hcWmlgMEQDJvDFLh8DA+3a2Mci8iaw0k5g8DcR+UJEvhGRn4K1kkREHrVzGb4PdG40JCIfichJ9usLRGSpne9vgZ1g4WfAL+3a5pn26pFX7Wt8ISKn2+fmiMh8OzfcNKwUWs0iIq/byQtWHJjAQET+bh9fICKd7GMDRGSufc7HIjLUlbtp6FDE1MoNg4Vds7sQa2E6WOtUj1HVjXbw2KOqJ4tIAvBfEZmPlR1mCDAMazXFSuDJA+x2Av4NnGXbylbV3SLyL6BCVR+wy70A/F1VP7GXSs3DSpE0BfhEVe8XkYuBax38Oz+xr5EEfCEir6pqMZACLFbVX4rIfbbtm7BWPPxMVdeJyEjgceCcVtxGQwfGBL7YIqnJ8rePsdYCnwZ8rqob7eNjgeMa+++ADGAQVv66maoaBLaLyAeHsD8KWNhoS1UPl6NwDDDMXoYGkG5noTkLOx+cqr4jIiUO/qdfiMjl9utetq/FWKmxXrKPPwfMtq9xGjCrybUTHFzDYNgPE/hii2o7HdVe7ABQ2fQQcLOqzjug3EUu+uHDyvpccwhfHCMio7GC6KmqWiUiHwGHS8mu9nVLD7wHBkO4mD6+I495wA12SixEZLC98H4h8H27D7AbcPYhzv0MOEtE+tnnZtvHy7HS5zcyHytJAHa54fbLhcAP7GMXAlkt+JoBlNhBbyhWjbMRH9BYa/0BVhO6DNgoIhPsa4iIHN/CNQyGgzCB78hjGlb/3VKxxJP+H1bN/jVgnf3eM8D/DjxRVXcBk7CalV+zr6n5FnB54+AG8AvgJHvwZCX7Rpd/ixU4V2A1efNb8HUuEBCRVcCfsQJvI5VYSVaXY/Xh3W8fnwhca/u3AhekBwwdD5OdxWAwdDhMjc9gMHQ4TOAzGAwdDhP4DAZDh8MEPoPB0OEwgc9gMHQ4TOAzGAwdDhP4DAZDh+P/AxufbzxiSjooAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAT4AAAEGCAYAAAD8EfnwAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAA+00lEQVR4nO2deXxU5fX/32eW7CEhJCxhkVUQUZEiYl0KgvuC+q2trba2Vay2VSutttaFX22r7ferVqpVS7HuK4q4sSoiaiuyaGWHsG8hBBKyLzNzfn/cGwgIyUzm3mQued6v131l7p17zz155s6ZZz0fUVUMBoOhPeFrawcMBoOhtTGBz2AwtDtM4DMYDO0OE/gMBkO7wwQ+g8HQ7gi0tQONycwJaF73ZMftFi933qbB4EVqqKROayUeG+eNTtc9e8NRnbvkq9rZqnp+PPdzg4QKfHndk/njtOMdt/v0wGMct2kweJGF+kHcNor3hlk4u0dU5wa7rc+N+4YukFCBz2AweAElrJG2diIuTOAzGAwxoUAEby98MIHPYDDETART4zMYDO0IRak3TV2DwdCeUCBsmrrusOLZTNZOzQCFY6+s4PgflfPhL3Mp2xgEoK7cR1JmhHFv7aR8m583L8wnq08IgLyTavnmfXujvteEh7dw6thySosD/PTsgY79D8NHlXHjH3bg9ykzX87htce6GLsO2XXLVzeeBbeer7z8Om6ftIXsvBAozHihE9OfynPMflOYPr4mEJHzgUmAH5iiqn+O5rqStUHWTs3gkqmF+ILKnOs703N0NaMfKd5/zud/7khSxoHqdmavEOPe2tkiP+e8msPbT+dy+6StLbr+cPh8ys/v386dV/WleGeQR2es47PZWWxZl2LsxmnXLV/BnWfBDZsA4ZAw+b58CpalkZoe5rFZa1m6INORcmgKBcIez+rk2soNEfEDfwcuAAYD3xORwdFcW7o+SN6JdQRSFV8Aup5Sy+Y5afvfV4WNM9Poc3GlI74uX5hBeYmzvwEDT65ix6YkCrckE6r3Mf+tbE47b5+x64Bdt3wFd54FN2wC7C0KUrDM+l5UV/rZWpBCbrd6x+9zOCJRbomKm0vWRgAFqrpBVeuAV4Bx0VzY8dg6di1JpqbER6ha2LYglcpC//73dy1OJrVTmKzeof3HKrYFeOuybsy4pguFi9t+pUanrvXs3pG0f794Z9CRh9LYdc9XL9OlRx39hlSzemla8yfHiaKEo9wSFTebut2BxnX7bcCph54kIjcANwDk5lsPc3a/ECdcX8ac6zoTSFVyBtUhjUL0hnfT6duotpfWOcyVH24npWOE4uVJfPDzPC5/bwdJGYlb8AaDU6SkhblnyiaevDefqgp/8xfEiSrUe/yr1eZJClR1sqoOV9XhmTkH4vCxV1Zw6bRCLnxxF8lZETr0tn7RIyHYPDeNPhdW7T/XnwQpHa2Kde6QOjr0Cu0fBGkr9hQGycuv27+f262e4p3x+2TsuuerF/EHlHumbGLetI58OjO7le4qhKPcEhU3A992oGej/R72saio3mO5VrHDz+Y5afS9xKrh7fh3Cll960nvemCRdM1eHxF7t3xrgLJNATJ7hr5mszVZ82Ua3fvU0aVnLYFghFHjSvlsTpax64Bdt3z1HsqEh7aydV0K0ya3zmiudVeIaHRbouJmU3cRMEBE+mAFvKuA70d78Yc351FT6sMXgJET95LcwSrFjTPS6XvRwYMahYtS+OJvWfgCgE857fd7Sc6Ovmv1t49v5sTTKsjKCfHC4pU8/1AXZr/cKerrD0ckLPz9ru7c/9IGfH6Y80oOm9fGP9pm7LrnK7jzLLhhE+D4EZWMvbKEDStTeHzuGgCefqAbi+Z1iNt2cyRybS4axE2xIRG5EHgEazrLv1T1T02d3/eEdDXZWQwG91ioH1Cme+OKWsefmKSvvNc5qnNP7LV9iaoOj+d+buDqPD5VnQHMcPMeBoOhdVGgXtt8eCAuEnblhsFgSEwUIdz246JxYQKfwWCImYh6u4/PBD6DwRATVpICE/gMBkO7QgibPj6DwdCesDIwm8DnGMXLk12ZejJ7x5eO2wQ4L3+oK3YNhkRGVahTZ5bGici/gIuBIlUdYh/LAV4FegObgO+oaomICFa2pwuBKuBHqrrUvuZa4G7b7B9V9dmm7uvtsG0wGNqECBLVFgXPAIfKT/4W+EBVBwAf2PtgZXoaYG83AE/A/kA5ESsXwAhgooh0bOqmJvAZDIaYsAY3fFFtzdpSXQAcmjV4HNBQY3sWuKzR8efU4jMgW0S6AecBc1V1r6qWAHP5ejA9iIRq6hoMBi8Q0+BGrogsbrQ/WVUnN3NNF1VtyCpcCDSk1z5cxqfuTRw/IibwGQyGmIhxcKM4niVrqqoi4vi6WtPUNRgMMRNWiWprIbvsJiz23yL7+JEyPsWcCcoEPoPBEBOKUK+BqLYW8jZwrf36WuCtRsd/KBYjgX12k3g2cK6IdLQHNc61jx0RzzV1W6Ku9Z0Tjic7N8TkD63UPQveyeL5h7qydV0Kf5uxlmNPqgZgyUcZ/Ov+fEL1QiCojL9nB0PPqABg/lvZvPK3LoTDcOrYMq6/u3lhI7fUtbykCuclX92yG0yO8NC0AoJJij+gfPxeNs8/2NUBb90rh6ZoGNxwAhF5GRiF1Re4DWt09s/AayJyHbAZ+I59+gysqSwFWNNZfgygqntF5A9YqfAA7lPVJmUW3RQb+peIFInIcqdsNqhr3X11H8aPGsjocaX0GlDT7HV/enHDQfu9B9Vw75RNnDDy4Lx+WTlh7nt2A/+Yt4bbJ23hf2/pBUDZXj9T/pDPn18r4J/z11CyO8gXH2c0e985r+Zw19V9YvgPo8MNuy0t2+bwkq9u2a2vFe64sh83nTOQm84ZyPBR5QwaFr9Qllv+NocSXTM3mqauqn5PVbupalBVe6jqU6q6R1XHqOoAVR3bEMTs0dyfq2o/VT1BVRc3svMvVe1vb083d183m7rP0MyQcqy0VF0rs2P4oP1eA2rp2b/2a+f1P6GaTl2tzM3HDKyhtsZHXa2wc0sS3fvWkt3JsnPymeV8MiO72fu6pa7lJVU4L/nqnnqbUFNlTfgNBBV/UHEiDaabanPNEcEX1ZaouObZEebnxEVrqmt98l4W/YdUk5Ss5PeuY9v6ZAq3JhEOwb9nZbF7+9Gl8eAl5TIvKcI14PMpj89dw6tfreCLBRms+SI9bptt9ZmpQlh9UW2JSpv38TVWWUvBfWm8aNi0JoWn/pTP/S+vByAzO8zND2zj/huPweeD44ZXsnNT20tYGrxDJCL87JyBpHcIM/GpjRwzsJrNa1Lb2q0WYQ1uuK/m5iZtHvjsyYyTATpITpMNgNZQ19q9I8h91/Xm9klbyO994F4jzy1j5LllAMx4oRN+nwJHj8CNl5TLvKQIdyiVZX7+++8MThldHnfga8vPzOuJSD3lvdvqWhX7/Nzzw7785Hc7OX7EwZ3PpcXWb0R5qZ93nsnl/O872opvc7ykXOYlRTiArJwQ6R2s/uGklAjDzqpga0H84kht9ZkpQkSj2xKVNq/xxUJL1bVuu2QA+/YGuPobg/nBrwrJ7Bjm8bu7s29PgHt+0Jd+x1dz/8sbePvpXHZsTOLFh7vy4sPWdIMHXllPdm6IJ+7pzoaV1i/01bcV0qPf1wdHDsUtdS0vqcJ5yVe37OZ0qefXk7bg84HPZ02nWvh+/EpobqrNNYfXa3yuqaw1np8D7AImqupTTV3TQXL0VBnjuC8mLZXBYOGEylrPIR10wtSRUZ07YfDc9qWypqrfc8u2wWBoS8SknjcYDO0LS17SjOoaDIZ2hKoQSeA5etFgAp/BYIiZRJ6cHA0m8BkMhpiw8vGZPj6DwdCuMPKSnsCtaSd3rv/KcZsP9DvRcZuGRoiHaiouTTWLF4WEnpwcDe0i8BkMBucwa3UNBkO7JJFTTkWDCXwGgyEmrLRUpqlrMBjaGaaPz2AwtCus7CymqWswGNoR1pI1E/hajURTq/rn+ccyftZaAKpL/Uy/pRf7tiWR1aOOyx7dQmpWmD3rk3n3Nz3YtSKVb00o5NTxxQDs2ZDMdFvMCKB0axJn/nKXa742h1tqaOAtf5/9bAXVFX4iEQiHhJsvdMa2W3bbQmUNU+M7MiLSE3gO6IL1IzFZVSfFY7NBraqmyo8/oDw8vYBF8zJZvTQ+/YIGtao7r+pL8c4gj85Yx2ezs9iyLvrcZv95Mo/e36zgtBt3858n8/jsyTxG/6aQlKwQ59y7g3VzDs6/1qlvLde9uw6ASBge++ZxDDx3Hx/8Md91Xw/HnFdzePvpXG6ftDUuO4fiNX8B7riyP2UuiEQ5bdetso0Gr6/ccDNsh4BfqepgYCTwcxEZHJ/JxFWrWvd+FidcUQLACVeUsHaulQk3PTdM/onV+JrICL7p3xlk96ojq3vzQjFeUkMD7/nrJdpKZa1hVNcJecm2wk2VtZ2qutR+XQ6sArrHazdR1aoqiwNkdLakKdPzQlQWR/+lXPVuNoMvKW01X1sTr/mLCve/vJ7HZq7hgquLE9puW5ZtRH1RbYlKq/xkikhv4GRg4WHei0llzQtqVSIgEl1VNFwnrPugA6NuL3TZK0M0TLi8P3sKk8jqVM+fX1nP1oIUli9sXjy+rey2BQ2aG17G9ZAsIhnAG8AvVbXs0PdVdbKqDlfV4UGil2xsrFYVL06oVaXnhqgosn5HKooCpHUKN3OFxfqPMulyfDXpuaFW87U18Z6/Vg1q354gn87MYtDQqoS121Zlq0BIfVFtiYqrnolIECvovaiq0+K1l8hqVQPGlLFsWkcAlk3ryICx0fW1rHwnm+OjbOY65Wtr4iV/k1PDpKaH97/+xrfK2bQm/ufLLbttWbamqXsERESAp4BVqvqwEzYTTa1q78ZkHjt9EGfeuouRNxYx/eZe/Pe1HLK6W9NZACp2B3jmsv7UVvgRgUXP5DJ+1lqSMyPUVQkbP83g/D9tc93X5nBLEc5L/nbMCzHxqY0A+P3w4fRsFs+P//lyy26bqawluHRkNLipsnYG8DGwDIjYh3+nqjOOdI1bKmtuYdJSeZB2npbKCZW1joM669n/+nZU5047/Yl2p7L2CXh8so/BYDgsXq/xJW4j3GAwJCQNiUij2ZpDRG4TkRUislxEXhaRFBHpIyILRaRARF4VkST73GR7v8B+v3dL/wcT+AwGQ0woQijii2prChHpDtwCDFfVIYAfuAr4C/BXVe0PlADX2ZdcB5TYx/9qn9ciTOAzGAwxE0Gi2qIgAKSKSABIA3YCZwOv2+8/C1xmvx5n72O/P8YeRI0ZE/gMBkNsaExN3VwRWdxou2G/GdXtwIPAFqyAtw9YApSqasPE1m0cWPHVHdhqXxuyz2/RUH77XuwYJw/0P8lxmzetW+e4TYAnBvR3xa7nSFABHy8Ro9hQ8ZFGdUWkI1Ytrg9QCkwFznfAxWYxgc9gMMSMQ6O6Y4GNqrobQESmAacD2SISsGt1PYDt9vnbgZ7ANrtpnAXsacmNTVPXYDDEhCKEI76otmbYAowUkTS7r24MsBL4EGiYKHgt8Jb9+m17H/v9edrCicimxmcwGGLGiXx8qrpQRF4HlmKlsfsCmAy8B7wiIn+0jz1lX/IU8LyIFAB7sUaAW4QJfAaDISZUnZvArKoTgYmHHN4AjDjMuTXAlU7c1wQ+g8EQM+rxlRsm8BkMhhjxfpICTwU+rwni9OhXw++e2LR/v2uvOp5/sCtvTukc1fVfPZPFytc6gMJx3ynjpB/vo3hVEgvu7Ux9lZDZPcTYhwpJylRqSnzMvrkrRctSGHRFGWdOjD3Lrxtl4JZAlJd8dctuXn4dt0/aQnZeCBRmvNCJ6U/lxW03GkyN7wiISAqwAEi27/O63Z5vMV4TxNm2PoWfnTto/z1eXLKCT2dmR3XtnrVJrHytA//zxjb8QeXd6/LpfXYV8+/qzDd/U0z+qTWsmprJl1M6MuK2vfiTlRG/3MvetUnsXZfU/A0Owa0ycEMgyku+umk3HBIm35dPwbI0UtPDPDZrLUsXZLouNqQK4Yi3A5+b01lqgbNV9SRgKHC+iIyMx6DXBHEaM/SMcnZuTqZoe3RBqXR9kC4n1RJMVXwByD+lmg2z09m3MUi3ETUA9Dyjmg2zrfTlwTSl2/Aa/Mktm6DrXhk4LxDlJV/dtLu3KEjBMkuuobrSz9aClNbT3HBuyVqb4KbYkKpqhb0btLeEnDbfGqIto8aVMn96dtTn5wyoY+fiFGpKfNRXC1s+SqeiMEDHAXVset+qKayfmUFFoTM/BG6WgdMCUV7y1W27DXTpUUe/IdWsXtq8bk28KFZTN5otUXE79bxfRL4EioC5qnpYsaGGdXz11LrpTpsRCEYYee4+FrybHfU1HfvXc/INJbzz43ze+0k+nY6rRXww+oEilr+YxdTLelBXKfiCCflbchANAlFXf2MwA4dWcczA6rZ26Yi45aubZZCSFuaeKZt48t58qir8jtk9MtGt003kARBXBzdUNQwMFZFs4E0RGaKqyw85ZzLWpEU6SE6bfIvdFm05ZXQ5BcvSKC2OzeZxV5Zz3JWWmNJnD+WQ0TVEx371XPLMDgBKNwbZMt+ZmkNrCNc0FoiKRxnPS766bdcfUO6Zsol50zpG3X/sBF5f8twqS9ZUtRRrGUqrLECOFbdFW0ZdVhJTM7eBqj3Wr3f5jgAb52Qw4JKK/cc0Akse78jgq5zpi3SrDNwQiPKSr27aBWXCQ1vZui6FaZNbZzR3/5093tR1c1Q3D6hX1VIRSQXOIY7EgeA9QRywVLWGnVXOpN/0jPna2b/oSm2JH19QOXPibpI7RPjqmSyWv2h9yfueW8mgbx+Q13xh1DHUVfgI1wsb52Zw8dPbyRkQXd+XW2XghkCUl3x10+7xIyoZe2UJG1am8PjcNQA8/UA3Fs2L33ZTWKO63l7m76bY0IlYSQP9WDXL11T1vqau8ZrYkBvCNTetNWmpDO7hhNhQav987fPgDc2fCKy6/PftTmzoK+Bkt+wbDIa2I5GbsdHgqZUbBoOh7VESu/8uGkzgMxgMMePxQV0T+AwGQ4woqMeXrJnAZzAYYsY0dQ0GQ7vD6xOYjxj4RORRmmjKq+otrnjkJVz49N2adnL16m2u2H1xUA9X7BoSl4a1ul6mqRrf4lbzwmAweAcFjtbAp6rPNt4XkTRVrXLfJYPBkOh4vanb7LoTETlNRFYCq+39k0Tkcdc9MxgMCYqgkei2RCWaBXePAOdhC/eq6n+Bs1z0yWAwJDoa5ZagRDWqq6pb5eB1qWF33DEYDAmPHt2DGw1sFZFvAioiQeBWYJW7bhkMhoQmgWtz0RBN4LsRmAR0B3YAs4Gfu+nUkXBLZc1rduNRGFv9XAYFU9NBof+VlQy61lIHWPN8OmtfykD8kP+tGobdvo+N76Sy6qnM/deWrAlywbQico6LLc27G4poYKVzf3TWWvbsDHLvtX3jtueWGpqb6oBOl0H0eLvG12wfn6oWq+rVqtpFVfNU9RpV3RPtDez081+IyLvxuWqprN11dZ94zXjaboPC2N1X92H8qIGMHldKrwE1UV1bujZAwdR0zn+tiAun72L7/BTKN/sp/CyZbfNSufCtXVz87i4G/8TK8dfnkmounF7EhdOLOO0ve8noEY456MXjb3Ncdn0xWx1UFGtQQ7vpnIHcdM5Aho8qZ9CwyrjtuvV8gfNlEDWRKLcEJZpR3b4i8o6I7BaRIhF5S0Ri+WlxrGnslsqal+zGozC2b0OQ3BPrCNjKbZ1PqWXL3FTWvZLO4PHl+G39npROX39iN7+XxjEXxj6byS1FtNxudYwYU8bMl3LitnUAd9TQ3Hq+3CmDKGiYxxfNlqBEM6r7EvAa0A3IB6YCL0djXER6ABcBU1rqoOFg4lEYyx5QT9HiJGpLfISqhR0fpVC1M0D5pgC7Fycz6zudmXtNHnuWfV2/YvPMNHpfFHvgc0sR7cbf72DKH7s5PmXCbTU0J3GrDKJBNbotUYkm8KWp6vOqGrK3F4Bo69aPAHfQRKW3PaisJQpZ/UIMHl/OvOtymTc+l47H1SN+JRIWavf5OO/VIk6+o5SPf9npoIe2+L9J+FOU7GNDbed8I04dW0ZpcWC/pqyTeEURzs0yiAqPT2c5YuATkRwRyQFmishvRaS3iBwjIncAM5ozLCIXA0WquqSp81R1sqoOV9XhQZJj/gfaG/EqjPX/dhUXTCvi3Bd2k9QhQofeIdK6hOl5TjUikHtiPeKD2pIDj8bmGaktqu054e/hGHxKJSPPLePZhSu584nNnHRGBXc8ujkum4fSWA0tEWmNMmgSh5q6IpItIq+LyGoRWWUvmMgRkbkiss7+29E+V0TkbyJSICJficiwlrrfVI1vCdZ63e8AP8VSSZsP3AR8NwrbpwOXisgm4BXgbBF5oaWOGiziVRir2WN95JU7/Gydm0rvi6voMbaaXZ9bPzplGwNE6iG5o1VJ14jVzD2mhYHPDUW0px/oxjXDB3PtqYN54KZj+O8nGfzvzcfEZRPcVENzHrfKIFpEo9uiYBIwS1UHASdhjQf8FvhAVQcAH9j7ABcAA+ztBuCJlvrf1FrduIahVPVO4E4AERkF/FpVr4nHplsqa16yG6/C2IJbOlFb6sMXUE65t5SkDkq/Kyr57K6OvHtJF3xB5bQ/l+zXUSpalExatxCZPVs2Z91NBTuncUsNza3nq81QAQf6FUUkC2sV2I8AVLUOqBORccAo+7RnsSpcvwHGAc+ppZD2mV1b7KaqO2O+dzQqayIyBBhMo749VX0u6pscCHwXN3We51TWPIRJS2UAZ1TWko/pqd3uvDWqczffdPtmoLjRocmqOhlARIYCk4GVWLW9JVizQLararZ9jgAlqpptT4n7s6p+Yr/3AfAbVY05k1SzY+wiMhEr+g7G6tu7APgEiDrwqep8rKhtMBiOBqIfuChuQl4yAAwDblbVhSIyiQPNWus2qioSZaM5BqIZ1f02MAYoVNUfY0Xm+GXrDQaDd3FmVHcbsE1VF9r7r2MFwl0i0g3A/ltkv78d6Nno+h72sZiJJvBVq2oECIlIB9uJns1cYzAYjlYcmsCsqoVYuQAa1vGNwWr2vg1cax+7FnjLfv028EN7dHcksK8l/XsQ3VrdxSKSDfwTqw1eAfynJTczGAxHBw42Pm8GXhSRJGAD8GOsCtlrInIdsBlrZglYXW0XAgVAlX1ui2g28Knqz+yXT4rILKCDqn7V0hsaDIajAIcCn6p+CRyuD/Bro5z2aK4jCVKaEhs64uRAERmmqkudcMBgMHgP54cbWpemanwPNfGeAmc77IsBQOKfH3U43Jp20m+RO3Py1o9wafliIi8g9RIJnIAgGpqawDy6NR0xGAweIcHX4UaDERQ3GAyxYwKfwWBob0gCJxmNBhP4DAZD7Hi8xhdNBmYRkWtE5F57v5eIjHDfNYPBkIhEm5klkUd+o1m58ThwGvA9e78c+LtrHhkMhsTH46nno2nqnqqqw0TkCwBVLbFnWbc6efl13D5pC9l5IVCY8UInpj+VF7fdRFRDOxI9+tXwuyc27d/v2quO5x/syptTOsdlN96yLX0pRNn0MAgk9xfy7g0iSbD3iRCVH0TABx3+x0/2VQHCFUrRPfWEdikaguxr/HS4NPZel2c/W0F1hZ9IBMIh4eYL4//s3FKEA++owkVFAtfmoiGap61eRPzY/6qI5BGlfpKdhLQcS4A81ESWhqgIh4TJ9+VTsCyN1PQwj81ay9IFmWyJU2Vqzqs5vP10LrdP2hqXncY0qIvdeVVfincGeXTGOj6bnRW3r9vWp/Czcwftv8eLS1bw6czsuP2Np2xDRcq+V8P0fDUJX4pQeGcdFXPCoBDapfScmoT4hNBe69tSNjVMUl+h21+TCJcoW75dS+YFfiQYew3hjiv7U+aQkI9bn1kDDYpoaRkty214KA2qcDVVfvwB5eHpBSyal8nqpe7rhCRyMzYaomnq/g14E+gsIn/CSkl1fwz3GK2qQ+MNegB7i4L7NQaqK/1sLUhxRLgm0dTQomXoGeXs3JxM0fb4K+Dxlq2GFK21/9ZAIE8oeyNMzvUBxGcFtEDOgcAWqQRVJVKl+DoI+OP+F+LGzc/MS6pwzaLWqG40W6ISzVrdF0VkCdbaOQEuU1VH5CLjoUuPOvoNqWb10jYSW2mGw6mLDRrWsvTtR2LUuFLmT8921CbEXraBzkL2NQE2X1KLJEPaqT7SRvrZdXc9FXMjVM6vx5ct5P46QFIvH1nf8bPzV/VsvqCWSBV0uT+4PzjGhAr3v7weFN57oRMzX8yN3UYj3PzMGhTR0jKcjQY+n/LY7LXk967jnWc6tZ4q3NFe4xORXliZEN7BSgtTaR+LBgXmiMgSEbnhCPZjVllLSQtzz5RNPHlvPlUVCVBVaAMCwQgjz93HgnezHbXbkrINlymVCyIc81YyvWcmE6mB8hlhtA4kCXo8l0yHy/zs/oNVg6z6LELyscIxM5Pp+WISxf8XIlIR+zdpwuX9+cX5A7nrmr5c+qNihpxaEbON1uCoVIXzuMpaNO2797D+BcFKPd8HWAMcH8W1Z6jqdhHpDMwVkdWquqDxCXYa6slgpZ5vzqA/oNwzZRPzpnV0pG/LLdxQF2vMKaPLKViWRmmxczZbWrbVn0cI5gv+jlatLWO0n5qvIgQ6C+mjrd/W9NE+dt9nfbzl74TJvjaAiBDsKQTyhbrNSsrxsdX69hRatbN9e4J8OjOLQUOrWL4wIyYbB9tz5zNrUEQ7ZcxKkpKVtMwwdzy62VFxoMaqcJvXpDpm90gc9X18qnqCqp5o/x0AjCDKfHyqut3+W4TVTxjn/D9lwkNb2bouhWmT4x/NdRM31MUaM+qyEoebuS0v20BXoWZZhEiNoqpULQqT1EdI/5aP6iVW065maYRgL9l/fvUiq4M/tEep3xwh2D22oJecGiY1Pbz/9Te+Vc6mNfENQrj1mRlVuMQj5h59VV0qIqc2d56IpAM+VS23X58L3NcCH/dz/IhKxl5ZwoaVKTw+dw1gPVSL5sWnhJWIamhNkZwaZthZ5Uz6jXOJsOMp25QhPjLG+Nh2TR34IXmg0OFyP5FaKLqnnn0v1SJpkHe3VXvqeF2Aot/Xs/WqWlSh0y+C+LNjC3wd80JMfGojAH4/fDg9m8Xz43sOvKQIB+6pwkWFx2t8zaqsiciERrs+rJz4nVT1vGau64tVywMrwL6kqn9q6hqjsoZraancGu4zaam8hRMqayn5PbX3DROaPxFY8/sJS5yY0eE00dT4Mhu9DmH1+b3R3EWqugFLmMhgMBxtePz3o8nAZ09czlTVX7eSPwaDIcERvD+40VTq+YCqhkTk9NZ0yGAweICjNfABn2P1530pIm8DU4HKhjdVdZrLvhkMhkQkwTOvREM0fXwpwB4sjY2G+XwKmMBnMLRXEng5WjQ0Ffg62yO6yzkQ8BrweLw3GAzxcDTX+PxABgcHvAY8/m8nMB6bbrH+lBpX7F68osQVu+8e39EVu+0Obz2mX6OpwLdTVeOacGwwGI5CEnwdbjQ0FfgSN32qwWBoU47mpm47X0JhMBiOyNEa+FR1b2s6YjAYvEMiJxmNBiMvaTAYYuMo7+MzGAyGryF4fwAgGs2NhMPnU/4+Zw33PbvBEXt5+XX879QCJs9fzeQPV3PZdbsdsTvh4S28+tUK/jFvjSP2Ghg+qowpH6/m6U9X8Z1f7HLUttNlG28ZbHg+mY/GdWD+pR3Y8FzyQe+tfyaZd4/vSF2J9TXc9m4SH12eyUeXdeDTqzMpWx17dm4vlS2462+TOJiBWUT8IvKFiLxr7/cRkYUiUiAirzaoOopIsr1fYL/fu6Xuuxr4RCRbRF4XkdUiskpETnPCboNalVM0KIzdMGoQt148gEt+VEyvAfHPT5vzag53Xd3HAQ8P0KAEdvfVfRg/aiCjx5U64msDTpdtPGVQts7H1teTOeOVMs6aVkbRR0EqN1uPbPVOofjTIKndDiiWpXUPc9ozFXxrehkDbqzmq/8XW6p3r5Wt2/42hcOC4rcCjXV8/gL8VVX7AyXAdfbx64AS+/hf7fNahNs1vknALFUdhJWiKm6RIjfUqox6m4UbZRtPGVRs8JN9Ygh/KvgCkDM8xM73rWSmK/6SxnG/qj6ozZVzcpikLOvbln1imOpdsT3eXivb1lDyOyIO1fhEpAdwETDF3hes5bGv26c8C1xmvx5n72O/P8Y+P2ZcC3wikgWcBTwFoKp1qloar90GtSqNuNPL4EX1NieCNLhftrGS2T/M3iUB6kqFcDUUfRykptBH4bwgKV0idBh0ZH3ardOS6HxmbOXitbJ1098miU1eMrdBTMzeDhUdewS4gwOrfzsBpaoasve3Ad3t192BrQD2+/vs82PGzRpfH2A38LTdfp9ip6A/iFhU1txUq4L2rd7mdtm2hMx+EfpdV8PC8Rks/GkGWYPCROqEgskpDPzFkdXEihcG2DotmeMmtJLiWDMkYtnGTfQ1vmJVHd5om9xgQkQuBopUdUkre+/qqG4AK63Vzaq6UEQmAb8F7ml8Uiwqa26qVbV39bbWUAJrCb3+p45e/2P9v6sfSSGpk1I4L8iCKyxtiZpdPhZ8uwNnvFJGSp5StsbPVxPTGPFkBUnZsc258FrZuq3k1xQOrdw4HbhURC7EygLVAat7LLshHyjQA9hun78d6AlsE5EAkIWVOSpm3KzxbQO2qepCe/91rEDYYtxSqzLqbW6WbXzU7rGahtU7hJ3vJ9FzXB3nfryPMXPLGDO3jJQuEc563Qp61TuExbemM/SBSjJ6xz7D1mtl67aSX5M40Menqneqag9V7Q1cBcxT1auBD4Fv26ddC7xlv37b3sd+f542Jxp0BFyr8alqoYhsFZGBqroGawncSrfuFw9Gvc094i2DJb9Mp67UhwSUE+6uItjhyM/52idTqd8nLP+D1aSUAJz5WnnU9/Ja2balvy6v1f0N8IqI/BH4AnucwP77vIgUAHuxgmWLaFZlLR5EZCjWaE0SsAH4saoeMd+QUVkzNGDSUrmDEypraXk9ddAV0amsfTHZuyprLUZVvwQS7p82GAwt56gWGzIYDIYjYgKfwWBob4jHMoUfigl8BoMhNkx2FoPB0B4xfXwGg6HdYRKReoGWrWNuGzzed+IUbk07ua0g7jwZX+OvAwY7btM1nHq8PP6Yto/AZzAYnCO2lFMJiQl8BoMhdkzgMxgM7QkzgdlgMLRLJOLtyGcCn8FgiA0zj691mfDwFk4dW05pcYCfnj3QUduXjy/igu/tRRU2rk7hoQm9qK+NL2tXeocQtz24ld4Da1CFh3/Vi1VLvpaLNWaGjyrjxj/swO9TZr6cw2uPdYnbptfsttTmcxf04YczNwJQU+rjvVu7U7YtiQ496rjob9tJyYqgCvP/0IWN8zMIpkY49y876TKkhrLtAd65qSeqEK4Xhv5wLyd9vzSq+z772QqqK/xEIpbGy80XOvP8umW3Ocx0liMgIgOBVxsd6gvcq6qPtNTmnFdzePvpXG6ftDVe9w6iU9c6LvtJMeNHD6KuxsddT25i1LgS5r4WXwqpm+7bzuIPO/DHG/oQCEZITo3/aWkQmLnzqr4U7wzy6Ix1fDY7iy1xith4ya5TNj//Ry49T6tixI1b+fzJTiz6RyfOvGM3mz5Kp3RTEj/+YD2FX6Ywb2JXvvfGJtLzQnx36iYCyUpdpfD8hX3pN6Yi6vvdcWV/yhzWYHHTbpN4vMbnWiJSVV2jqkNVdSjwDaAKeDMem26I9zTgDyjJKRF8fiU5NcKewvgy2aZlhjnh1EpmvWyJy4TqfVSWxe+7WwIzXrLrlM0N72cw+ArrusFX7GP93EwA1r+fyXGX70MEup1cQ22Zj4qiAP4kCCRb3/hwnS9htEnaAodV1lqd1vqZGAOsV9XNrXS/mNhTmMTrT3bm+c9XUlsjLP2oA0sXxJeEtGuvWvbtCfCrv26h7+Aa1n2VyhP3dqe2Oj4tj8MJzAwaVhWXTa/ZdcpmVXGAjM6Wpk16XoiqYuvrULErQGYj0Z6MriEqdlnnlu8IMH18T0o3J3Hmb4rI6BI6rO2vocL9L68Hhfde6MTMF3Nj9rdV7TZ5Tzw/0b61BMWvAl4+3BuxiA25RUZWiNPO28e1Iwfz/WFDSEkLc/YVe+Oy6fdD/xOqePe5XH5+3kBqqnx89xdFDnlscBoRDpKqPBKZ+SF+8N5GfvzBela+mUVlcXQ/ZBMu788vzh/IXdf05dIfFTPk1OibyG1htzliUFlLSFwPfLYK+qXA1MO9r6qTGxSYgiS77c5hOfnMCgq3JLFvb4BwSPh0ZjaDh1fGZbN4Z5DdO4Os+cIazPjkvWz6nxC/6pdbAjNesuuUzbTcEBVFdi2vKEBaJ6v2ltElRHkjexWFga/V7DK6hMg9tpbti6JTTttTaNVQ9+0J8unMLAYNjb827abdpmiYx+flpm5r1PguAJaq6q5WuFeLKNoe5LhhVSSnRABl6BnlcXfql+wOUrwjiR79LGX7oWeUs2Vt/IHdLYEZL9l1ymbfMRWsnGZdt3JaFn3HVuw/vurNLFRh5xcpJGVGrGbuzgChGqtaWLPPx/bFqeT0rTui/QaSU8Okpof3v/7Gt8rZtCZ+bQy37DaLavRbgtIafXzf4wjN3FhxQ7wHYM0X6Xz8XhZ/n72GcEgoWJHKzBfjt/v3e7rzm0c3EwgqhVuSeGhCr7htuiUw4yW78dgs2ZjMP0/vz2m37uaUn+7hvVu6s2JqNpnd67n4b9sA6DOqgk3z03n67H4E7OksAHvXJ7Pggc5WlUfhG9fvJXdg890zHfNCTHzKmkLj98OH07NZPD++PmQ37UZDItfmosFtsaF0YAvQV1WbHXZzTWzIZGcx2LT37CwLI+/HLTaUmd1DTz7r1qjO/fidO9ql2FAlEH/VyWAwJBRer/F5auWGwWBIABQIezvymcBnMBhixtT4DAZD+8PjfdEm8BkMhpgxNT6DwdC+MGmpPILHq+UG5/hr/+Mct/mdVTsdtwnw2nFdXbEbLwKIGdwwGAztDfF4ZaK1khQYDIajBY1hawIR6SkiH4rIShFZISK32sdzRGSuiKyz/3a0j4uI/E1ECkTkKxEZ1tJ/wQQ+g8EQI46t1Q0Bv1LVwcBI4OciMhj4LfCBqg4APrD3wVr3P8DebgCeaOl/YAKfwWCIGSeys6jqTlVdar8uB1YB3YFxwLP2ac8Cl9mvxwHPqcVnQLaIdGuJ/6aPz2AwxE70fXy5IrK40f5kVZ186Eki0hs4GVgIdFHVhhGjQqBBUKU70Fh3Ypt9LObRJRP4DAZDbGhMo7rFzSUpEJEM4A3gl6paJo2Siqiqijg/a9BTgS8vv47bJ20hOy8ECjNe6MT0p/Icse2GEphbqnBuqs35fMqjs9ayZ2eQe6/t64hNr5StE8/X2ufS2DA1FRT6XlnNsddaiUHXvZBGwUtpiE/p9q1aTrrdyv23anI6G99IRXxw8l1ldD2j+fx+DQSTIzw0rYBgkuIPKB+/l83zD7bSFBiHQpGIBLGC3ouqOs0+vEtEuqnqTrsp25C6fDvQs9HlPexjMeNqH5+I3GaP1iwXkZdFJK5EbOGQMPm+fG4YNYhbLx7AJT8qpteAmrj9bFDtuvvqPowfNZDR40odsTvn1RzuurpP3HZayy7AZdcXszXOJKyN8VLZxvt87VsbYMPUVMa+todzp+9hx/xkyjf7KVqYxPYPkjl3ejHnv7uHgT+xguG+Aj9bZqRw3jvFnPnPEpbc14FIOHp/62uFO67sx03nDOSmcwYyfFQ5g4bFlzk8WkQ1qq1JG1bV7ilglao+3Oitt4Fr7dfXAm81Ov5De3R3JLCvUZM4JlwLfCLSHbgFGK6qQwA/lvZGi9lbFKRgmZXqu7rSz9aCFHIbicK0FLcUxtxShXPLbm63OkaMKWPmSzmO2fRS2cb7fJVt8NPpxHoCqeALQN4pdWyfm0LBK6kcN74Sv62PlNLJEqPYMS+FXhfW4E+CjB5hMnqF2ftVLCn0hZoqS/MjEFT8QW29ufrOjOqeDvwAOFtEvrS3C4E/A+eIyDpgrL0PMAPYABQA/wR+1lL33W7qBoBUEakH0oAdThnu0qOOfkOqWb00Os2DpnBLYcxr3Pj7HUz5YzfSMpxTifFq2bbk+coaEGLZI5nUlgj+FKVwQTIdh9RTsSnA7iVJLJuUgT8JTrqjjJwTQlTv8tHppAOBNa1LmOqi2OoiPp/y2Oy15Peu451nOu3XeHEVBRx4RFT1E44s8fS1jMRqZU3+efx3dldXdzvwIFYG5p1Y1dI5h57XEpW1lLQw90zZxJP35lNVEZ9co8Hi1LFllBYH9td42jMtfb469Asz6PpKFlyfw4LxOWQPCiE+iISgbp8w5pW9nHh7Of+5LduxmlkkIvzsnIFc/Y3BDBxaxTED4xe0ag4humZuIq/ucLOp2xFr3k0fIB9IF5FrDj0vVpU1f0C5Z8om5k3ryKczsx3x1S2FMS8x+JRKRp5bxrMLV3LnE5s56YwK7ng0fhlkr5VtvM9X329Xc84bezj7hb0EsyJk9g6R1jVCj3NqEYFOJ9aDD2pLhNQuEaoKDwTWql1+Uju3rCpVWebnv//O4JTR5S26PmYikei2BMXNwY2xwEZV3a2q9cA04JvxmVQmPLSVretSmDbZmdFccE9hzEs8/UA3rhk+mGtPHcwDNx3Dfz/J4H9vPiZuu94q2/ifr5o91leqcoeP7XNT6HVxDfljaihaaDX3yzf6idQLyR2V/NG1bJmRQrgOKrb5qdjsJ+fE6PsUs3JCpHewRkOSUiIMO6uCrQWtobKG1dSNZktQ3Ozj2wKMFJE0oBqrzb646Uua5vgRlYy9soQNK1N4fO4awPrCLpoXn7KUWwpjbqnCuWXXDbxUtk48X/++NZu6Uh8SUIbdU0ZSB6XPFdUsujuLWZd0wheEEQ/sQ8TqE+x5fg2zLs7F54dh95Thi6HnJqdLPb+etAWfD3w+WPBOFgvfby2VtcRtxkaD2yprvwe+i7Um7wvgelU9YkeeayprBoOLfGdVoSt23UhLtVA/iFtlLSstX08bcF1U587+6o/tUmVtIjDRzXsYDIbWJrHFwqPBUys3DAZDAmBU1gwGQ3vE6318JvAZDIbYMYHPYDC0KxSImMBnMBjaFWZww+AVYpkgFguxpBNJBCSumRyHxS01tB+viX/lzKGsvyK6ZaHNYgKfwWBoVygQTuBlGVFgAp/BYIgRBTWBz2AwtDdMU9dgMLQrzKiuwWBol5gan8FgaHeYwNd6uKWy5pZalRvqYg04oYY24cHNnDp2n6VUNnYwAGdeVMIPJuyk54Aabrl4IOu+ii+VuVuKcG6V7eXji7jge3tRhY2rU3hoQi/qa51JW9nSz2zFs5msnZoBCsdeWcHxPyrnw1/mUrbRSuhaV+4jKTPCuLcs3Z29q4P8e2In6isEfHDJ6zsJNJ/jN3pUIeyxaUyH4GrgE5FbgfFYefX/qaqPxGOvQQWrYFkaqelhHpu1lqULMtkSpypYg1pVTZUff0B5eHoBi+Zlsnppy7/0Depid17Vl+KdQR6dsY7PZmfF7WsDDWpoaRktfwDnTM3h7WfyuP2RTfuPbVqTwn3j+3LLX7Y44KWlhvb207ncPmlr8ydHiVtl26lrHZf9pJjxowdRV+Pjric3MWpcCXNfcybXYUs+s5K1QdZOzeCSqYX4gsqc6zvTc3Q1ox8p3n/O53/uSJKtkxIJwYLbcznr/4rJGVRPTYkPnxvfco/X+NxMPT8EK+iNAE4CLhaR/vHYdEtlzQ21KrfUxcA5NbTlCzMpLz14YvPWglS2bXAui68bamhulq0/oCSnRPD5leTUCHsKnUmT39LPrHR9kLwT6wikKr4AdD2lls1zDuiiqMLGmWn0udiSldz+aQodB9aRM8j6XqR0jLgzd90ZlbU2w83U88cBC1W1SlVDwEfAFU4Zd1JlDaxaxONz1/DqVyv4YkFG3GpVh1MXcyZIH1BD04jzqxC8gFtlu6cwidef7Mzzn6/k5S+WU1nmZ+kCZzIat/Qz63hsHbuWJFNT4iNULWxbkEplI52OXYuTSe0UJqt3CICyjUFEYPZ1nXnr8q4s+6cbGZnVGtWNZktQ3Ax8y4EzRaSTnX7+Qg5WQQcSR2WtLdSqWoJRQ3OPjKwQp523j2tHDub7w4aQkhbm7Cv2xm03ns8su1+IE64vY851nZlzfWdyBtUhjb61G95Np+/FB0TEI2Fh15IUvvV/xVz00i42v5/Gjv84rMOhoBqJaktUXOvjU9VVIvIXYA5QCXwJfK1zQ1UnA5PBSj3fnF03VNYa01itavOa1BbbcUtdrEEN7ZQxK0lKVtIyw9zx6GZHhIG8gltle/KZFRRuSWLfXutr8enMbAYPr2TetPi6FOL9zI69soJjr6wAYMnD2aR1sWp3kRBsnpvGpdN27j83vWuILqfUkJJjBZ0eZ1WzZ0US+afVxPU/fA2PL1lzs8aHqj6lqt9Q1bOAEmBtnBZdUVlzQ63KLXUxt9TQvIRbZVu0Pchxw6pITokAytAzyh0ZjIr3M6u2ldsqdvjZPCeNvpdYNbwd/04hq2896V0P1Ce6n1FDydogoWohEoLCRclk93emi2U/qp6Xl3R7VLezqhaJSC+s/r2R8dhzS2XNDbUqt9TFnOS3j23kxNPKLaWyRct4/qFulJcG+NkftpKVE+IPz65n/YpU7rpmQMvv4YIamltlu+aLdD5+L4u/z15DOCQUrEhl5ottr1734c151JRao7MjJ+4luYPVMNo4I52+F1UedG5yVoQhPyrnnW93BbFqfD1HudBtk8ADF9Hgtsrax0AnoB6YoKofNHW+UVlzEZOWysKFtFRuBQE30lLdfcUKNiyrjE9lzZ+rI1MviurcOZXPtUuVtTPdtG8wGNqCxJ6qEg2eWrlhMBgSAJOkwGAwtDcUUI8vWXN1VNdgMByFqJ2INJqtGUTkfBFZIyIFIvLbVvAeMDU+g8HQAtSBpq6I+IG/A+cA24BFIvK2qq6M23gzmBqfwWCIHWdqfCOAAlXdoKp1wCvAONd9x+XpLLEiIruBaMbwc4HiZs+KHWPXW756zW4i+HqMqsY1+19EZtn3jIYUoPGykcn2ai1E5NvA+ap6vb3/A+BUVf1FPP5FQ0I1daP9QERksRtzg4xdb/nqNbte8rUpVPX81rqXW5imrsFgaCu2c3Dikh72Mdcxgc9gMLQVi4ABItJHRJKAq4C3W+PGCdXUjYHJxq5rdr3kq9fseslX11HVkIj8ApgN+IF/qeqK1rh3Qg1uGAwGQ2tgmroGg6HdYQKfwWBod3gu8LmxxEVE/iUiRSKy3Al7ts2eIvKhiKwUkRW24pwTdlNE5HMR+a9t9/dO2G1k3y8iX4jIuw7a3CQiy0TkSxFZ7JDNbBF5XURWi8gqETnNAZsDbR8btjIR+aUD7iIit9mf13IReVlEHEnOKCK32jZXOOVru0BVPbNhdYCuB/oCScB/gcEO2D0LGAYsd9DXbsAw+3UmVvZpJ3wVIMN+HQQWAiMd9HsC8BLwroM2NwG5Dj8LzwLX26+TgGwXnrVCrAm/8drqDmwEUu3914AfOWB3CJa2TRrWQOX7QH8ny+Fo3bxW43NliYuqLgDiV5U52OZOVV1qvy4HVmF9AeK1q6paYe8G7c2RESoR6QFcBExxwp5biEgW1o/VUwCqWqeqpQ7fZgywXlWdygYaAFJFJIAVqHY4YNNVJcOjGa8Fvu5AY2XqbTgQTNxGRHoDJ2PVzpyw5xeRL4EiYK6qOmIXeAS4A3BaLEGBOSKyRERucMBeH2A38LTdLJ8iIvHpgX6dq4CXnTCkqtuBB4EtwE5gn6rOccB0VEqGhq/jtcDnOUQkA3gD+KWqljlhU1XDqjoUa6b7CFu8PS5E5GKgSFWXxGvrMJyhqsOAC4Cfi8hZcdoLYHVNPKGqJ2Op+DmW0sieTHspMNUhex2xWiZ9gHwgXUSuideuqq4CGpQMZ3EEJUPD1/Fa4GuzJS4tQUSCWEHvRVWd5rR9u3n3IeDE2snTgUtFZBNWF8LZIvKCA3YbajyoahHwJlaXRTxsA7Y1qum+jhUIneICYKmq7nLI3lhgo6ruVtV6YBrwTScMq+NKhu0DrwW+NlviEisiIlh9UKtU9WEH7eaJSLb9OhUrl9nqeO2q6p2q2kNVe2OV6zxVjbtWIiLpIpLZ8Bo4F6uJFo+vhcBWERloHxoDOJnD7Xs41My12QKMFJE0+7kYg9XnGzci0tn+26Bk+JITdo92PLVkTV1a4iIiLwOjgFwR2QZMVNWn4jR7OvADYJndHwfwO1WdEafdbsCzdhJHH/Caqjo29cQFugBvWt93AsBLqjrLAbs3Ay/aP4AbgB87YLMhOJ8D/NQJewCqulBEXgeWAiHgC5xbZvaGiDQoGf7chUGeoxKzZM1gMLQ7vNbUNRgMhrgxgc9gMLQ7TOAzGAztDhP4DAZDu8MEPoPB0O4wgc9DiEjYzhqyXESm2suUWmrrGVvlCnvJ1+Amzh0lIjFPuLWzsnxNjetIxw85p6Kp9w9z/v8TkV/H6qOhfWICn7eoVtWhqjoEqANubPymvQA+ZlT1em1axHkUDq00MBgSARP4vMvHQH+7NvaxiLwNrLQTGPyfiCwSka9E5KdgrSQRkcfsXIbvA50bDInIfBEZbr8+X0SW2vn+PrATLNwI3GbXNs+0V4+8Yd9jkYicbl/bSUTm2LnhpmCl0GoSEZluJy9YcWgCAxH5q338AxHJs4/1E5FZ9jUfi8ggR0rT0K7w1MoNg4Vds7sAa2E6WOtUh6jqRjt47FPVU0QkGfhUROZgZYcZCAzGWk2xEvjXIXbzgH8CZ9m2clR1r4g8CVSo6oP2eS8Bf1XVT+ylUrOxUiRNBD5R1ftE5CLguij+nZ/Y90gFFonIG6q6B0gHFqvqbSJyr237F1grHm5U1XUicirwOHB2C4rR0I4xgc9bpDZa/vYx1lrgbwKfq+pG+/i5wIkN/XdAFjAAK3/dy6oaBnaIyLzD2B8JLGiwpapHylE4FhhsL0MD6GBnoTkLOx+cqr4nIiVR/E+3iMjl9uuetq97sFJjvWoffwGYZt/jm8DURvdOjuIeBsNBmMDnLartdFT7sQNAZeNDwM2qOvuQ8y500A8fVtbnmsP4EjUiMgoriJ6mqlUiMh84Ukp2te9bemgZGAyxYvr4jj5mAzfZKbEQkWPthfcLgO/afYDdgNGHufYz4CwR6WNfm2MfL8dKn9/AHKwkAdjnDbVfLgC+bx+7AOjYjK9ZQIkd9AZh1Tgb8AENtdbvYzWhy4CNInKlfQ8RkZOauYfB8DVM4Dv6mILVf7dULPGkf2DV7N8E1tnvPQf859ALVXU3cANWs/K/HGhqvgNc3jC4AdwCDLcHT1ZyYHT591iBcwVWk3dLM77OAgIisgr4M1bgbaASK8nqcqw+vPvs41cD19n+rcAB6QFD+8NkZzEYDO0OU+MzGAztDhP4DAZDu8MEPoPB0O4wgc9gMLQ7TOAzGAztDhP4DAZDu8MEPoPB0O74/x4HgpEGQjBZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.fit(\n",
    "    x_train_norm,\n",
    "    y_train_norm,\n",
    "    batch_size=batch_size,\n",
    "    epochs=epochs,\n",
    "    # validation_split=0.1,\n",
    "    callbacks=[tensorboard_callback, cm_callback],\n",
    "    validation_data=(x_test_norm, y_test_norm),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "34c52b84-d316-46d6-9dfc-0d7d02cc8cc8",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PosixPath('logs/mnist-fit/20220418-022041')"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log_dir"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "a3b310df-24b4-4125-ae14-090d32e6f262",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "      <iframe id=\"tensorboard-frame-1ed78cf4697c56ef\" width=\"100%\" height=\"800\" frameborder=\"0\">\n",
       "      </iframe>\n",
       "      <script>\n",
       "        (function() {\n",
       "          const frame = document.getElementById(\"tensorboard-frame-1ed78cf4697c56ef\");\n",
       "          const url = new URL(\"/\", window.location);\n",
       "          const port = 6008;\n",
       "          if (port) {\n",
       "            url.port = port;\n",
       "          }\n",
       "          frame.src = url;\n",
       "        })();\n",
       "      </script>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%tensorboard --logdir logs/mnist-fit/20220418-022041"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "114db0cf-0f12-4c93-86a2-08b8d1414335",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Avaliando o modelo\n",
    "\n",
    "Agora que treinamos o nosso modelo, podemos avaliar a sua real acurácia, introduzindo dados nunca antes vistos pelo modelo.\n",
    "Essa etapa é extremamente importante, para avaliarmos se o modelo durante o treino foi capaz de inferir as características que fazem\n",
    "o dígito 5, ser 5, ou um 4, ser 4, etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "99b7bc3b-bcd2-465d-beeb-afa68fb92e13",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 2s 6ms/step - loss: 0.1097 - accuracy: 0.9791\n",
      "Acurácia do modelo: 97.91%\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(x_test_norm, y_test_norm, verbose=1)\n",
    "print(f\"Acurácia do modelo final: {score[1]:.2%}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a1cd5bd7-6021-44cd-8047-89d5ca501b6b",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    ":::{note}\n",
    "\n",
    "Repare que a acurácia final que obtivemos durante o teste do modelo (97,91%) é ligeiramente menor do que o valor que obtivemos ao final do treino.\n",
    "Isso já era esperado, uma vez que essa é a primeira vez que o nosso modelo analisa esse conjunto de imagens.\n",
    "\n",
    ":::"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76ff4c72-0818-4259-bf32-cf0fd02c688c",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Salvando o modelo treinado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "32dbac27-02b9-48d5-9a33-ee28ff11d61b",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "model.save('models/mnist_feed-forward', save_format='h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "250b6838-5bd7-4646-8280-ee164cec8797",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "## Criando função `predict_image`\n",
    "\n",
    "A função `predict_image` utiliza o modelo treinado para prever uma imagem ou lista de imagens.\n",
    "Ela também utiliza a função `plot_imgs` criada anteriormente para visualizar as imagens."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 330,
   "id": "964caa51-4a68-43b7-ab39-a64402571205",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def predict_image(\n",
    "    img: Iterable | np.ndarray,\n",
    "    real_label: Iterable | str | int | None = None,\n",
    "    model: keras.engine.sequential.Sequential = None,\n",
    "    plot_results: bool = True\n",
    "):\n",
    "    \"\"\"\n",
    "    Predict the label of an image or array of images.\n",
    "\n",
    "    Function also normalizes images, if they're not normalized already.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    img : Iterable | np.ndarray\n",
    "        The image to predict.\n",
    "    real_label : Iterable | str | int | None, optional\n",
    "        The real label of the image.\n",
    "        If not None, the predicted label will be displayed next to the real one.\n",
    "    model : keras.engine.sequential.Sequential, optional\n",
    "        The model used to predict the image.\n",
    "        If None, a ValueError will be raised.\n",
    "    plot_results: bool, optional\n",
    "        Whether to plot the image(s).\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    preds : List[int]\n",
    "        The predicted label, or labels of the image(s).\n",
    "\n",
    "    Raises\n",
    "    ------\n",
    "    ValueError\n",
    "        If the model is None.\n",
    "\n",
    "    Examples\n",
    "    --------\n",
    "    You can use this function as follows:\n",
    "\n",
    "    .. code-block:: python\n",
    "\n",
    "        # predicts and plots the first five images from test dataset.\n",
    "        predict_img(x_test[0:5], y_test[0:5], model)\n",
    "\n",
    "        # predicts the first five images from test dataset.\n",
    "        predict_img(x_test[0:5], y_test[0:5], model, plot_results=False)\n",
    "\n",
    "        # predicts and plots the first five images from normalized test dataset.\n",
    "        predict_img(x_test_norm[0:5], y_test[0:5], model)\n",
    "\n",
    "    \"\"\"\n",
    "    if model is None:\n",
    "        raise ValueError('You need to pass the trained model to the \\\"model\\\" argument')\n",
    "\n",
    "    _img = img\n",
    "    if not img.ndim == 4:\n",
    "        img = img.reshape(-1, 28, 28, 1)\n",
    "    max_pixel = img.max()\n",
    "    if max_pixel > 1:\n",
    "        img = img.astype('float32')/255\n",
    "\n",
    "    preds = model.predict(img)\n",
    "    preds = [np.argmax(pred) for pred in preds]\n",
    "    if plot_results:\n",
    "        if real_label is not None:\n",
    "            labels = [f'Real: {real_lbl}/ Pred: {pred}' for real_lbl, pred in zip(real_label, preds)]\n",
    "        else:\n",
    "            labels = preds\n",
    "        plot_imgs(_img, labels)\n",
    "\n",
    "    return preds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 331,
   "id": "67cd3193-a7cf-4d76-a60f-52f5740296d7",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "[7, 2, 1, 0, 4]"
      ]
     },
     "execution_count": 331,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_image(x_test[0:5], y_test[0:5], model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e304287-c92c-4367-a4f6-f6bdbbed19a5",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Erros do modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 332,
   "id": "025d57f2-4bec-4dbf-899e-0f4e4bc38db5",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "predictions = predict_image(x_test, model=model, plot_results=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 333,
   "id": "d7272d66-c0f1-41e9-9bb7-c9c9ecefdbf6",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Modelo errou: 76 de 10,000 (0.76%)\n"
     ]
    }
   ],
   "source": [
    "false_predictions = [\n",
    "    idx for idx, (true_label, predicted_label) in enumerate(zip(y_test, predictions))\n",
    "    if true_label != predicted_label\n",
    "]\n",
    "print(\n",
    "    f'Modelo errou: {len(false_predictions):,} de {len(y_test):,} ({len(false_predictions)/len(y_test):.2%})'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 334,
   "id": "49802f93-d6f1-4241-a656-1323b88a9e51",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1296 with 30 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "false_preds = predict_img(x_test_norm[false_predictions], y_test[false_predictions], model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48eaf079-b15a-4356-8cb5-f45042642c81",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "## **Opção 2:** Tensorboard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "dc5eddba-17ae-4130-b8ab-d7f38e9cc0c3",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 1s 3ms/step - loss: 0.0816 - accuracy: 0.9847\n",
      "Acurácia do modelo: 98.47%\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(x_test_norm, y_test_norm, verbose=1, callbacks=[tensorboard_callback, cm_callback])\n",
    "print(f\"Acurácia do modelo: {score[1]:.2%}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "16f9095b-5cb3-4927-a313-3731fc05132d",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "model.save('outputs/models/mnist_example01')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6aa27412-f286-40ff-8da2-50fa15f94126",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Conclusão\n",
    "\n",
    "O modelo criado foi capaz de aprender com alta acurácia, a reconhecer dígitos escritos à mão.\n",
    "\n",
    "\n",
    "Como observado em [Erros do modelo](#Erros-do-modelo), as imagens que o nosso modelo não foi capaz de prever corretamente, seriam extremamente difíceis de serem classificadas até mesmo por um humano treinado. Isso porque, elas se assemelham bastante à outros dígitos. Sendo assim, podemos classificá-las como \"outliers\".\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.8"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}