Automated ci tests (#2)

.github/workflows/run-ci.yml

@@ -0,0 +1,25 @@
+name: Run CI
+
+on: [pull_request]
+
+jobs:
+ build:
+
+ runs-on: ubuntu-latest
+ strategy:
+ matrix:
+ python-version: ["3.9", "3.10", "3.11"]
+
+ steps:
+ - uses: actions/checkout@v2
+ - name: Set up Python ${{ matrix.python-version }}
+ uses: actions/setup-python@v2
+ with:
+ python-version: ${{ matrix.python-version }}
+ - name: Install dependencies
+ run: |
+ python -m pip install --upgrade pip
+ make install
+ - name: Run tests
+ run: |
+ make test

Makefile

@@ -0,0 +1,8 @@
+install:
+ pip install -e .[all]
+
+upgrade:
+ pip install --upgrade -e .[all]
+
+test:
+ python -m pytest

README.md

@@ -17,11 +17,7 @@ pip install pykalman-bardo
 Alternatively, you can setup from source:
 
 ```bash
-# pip
-pip install -e .
-
-# poetry
-poetry pyproject.toml
+pip install .
 ```
 
 ## Usage

pykalman/sqrt/tests/test_unscented.py

@@ -3,8 +3,6 @@
 from numpy.testing import assert_array_almost_equal
 from scipy import linalg
 
-from nose.tools import assert_true
-
 from pykalman.sqrt import AdditiveUnscentedKalmanFilter
 from pykalman.sqrt.unscented import cholupdate, qr
 
@@ -46,8 +44,8 @@ def test_additive_sample():
  kf = build_unscented_filter(AdditiveUnscentedKalmanFilter)
  (x, z) = kf.sample(100)
 
- assert_true(x.shape == (100, 2))
- assert_true(z.shape == (100, 1))
+ assert x.shape == (100, 2)
+ assert z.shape == (100, 1)
 
 
 def test_additive_filter():

pykalman/tests/test_standard.py

@@ -5,7 +5,6 @@
 import numpy as np
 from numpy.testing import assert_array_almost_equal
 from scipy import linalg
-from nose.tools import assert_true
 
 from pykalman import KalmanFilter
 from pykalman.datasets import load_robot
@@ -29,8 +28,8 @@ def test_kalman_sampling(self):
  self.data.initial_state_covariance)
 
  (x, z) = kf.sample(100)
- assert_true(x.shape == (100, self.data.transition_matrix.shape[0]))
- assert_true(z.shape == (100, self.data.observation_matrix.shape[0]))
+ assert x.shape == (100, self.data.transition_matrix.shape[0])
+ assert z.shape == (100, self.data.observation_matrix.shape[0])
 
  def test_kalman_filter_update(self):
  kf = self.KF(
@@ -128,7 +127,7 @@ def test_kalman_fit(self):
  loglikelihoods[i] = kf.loglikelihood(self.data.observations)
  kf.em(X=self.data.observations, n_iter=1)
 
- assert_true(np.allclose(loglikelihoods, self.data.loglikelihoods[:5]))
+ assert (np.allclose(loglikelihoods, self.data.loglikelihoods[:5])).all()
 
  # check that EM for all parameters is working
  kf.em_vars = 'all'
@@ -137,7 +136,7 @@ def test_kalman_fit(self):
  kf.em(X=self.data.observations[0:n_timesteps], n_iter=1)
  loglikelihoods[i] = kf.loglikelihood(self.data.observations[0:n_timesteps])
  for i in range(len(loglikelihoods) - 1):
- assert_true(loglikelihoods[i] < loglikelihoods[i + 1])
+ assert (loglikelihoods[i] < loglikelihoods[i + 1]).all()
 
  def test_kalman_initialize_parameters(self):
  self.check_dims(5, 1, {'transition_matrices': np.eye(5)})
@@ -154,16 +153,14 @@ def check_dims(self, n_dim_state, n_dim_obs, kwargs):
  initial_state_mean, initial_state_covariance) = (
  kf._initialize_parameters()
  )
- assert_true(transition_matrices.shape == (n_dim_state, n_dim_state))
- assert_true(transition_offsets.shape == (n_dim_state,))
- assert_true(transition_covariance.shape == (n_dim_state, n_dim_state))
- assert_true(observation_matrices.shape == (n_dim_obs, n_dim_state))
- assert_true(observation_offsets.shape == (n_dim_obs,))
- assert_true(observation_covariance.shape == (n_dim_obs, n_dim_obs))
- assert_true(initial_state_mean.shape == (n_dim_state,))
- assert_true(
- initial_state_covariance.shape == (n_dim_state, n_dim_state)
- )
+ assert transition_matrices.shape == (n_dim_state, n_dim_state)
+ assert transition_offsets.shape == (n_dim_state,)
+ assert transition_covariance.shape == (n_dim_state, n_dim_state)
+ assert observation_matrices.shape == (n_dim_obs, n_dim_state)
+ assert observation_offsets.shape == (n_dim_obs,)
+ assert observation_covariance.shape == (n_dim_obs, n_dim_obs)
+ assert initial_state_mean.shape == (n_dim_state,)
+ assert initial_state_covariance.shape == (n_dim_state, n_dim_state)
 
  def test_kalman_pickle(self):
  kf = self.KF(

pykalman/tests/test_unscented.py

@@ -4,8 +4,6 @@
 from numpy import ma
 from numpy.testing import assert_array_almost_equal
 
-from nose.tools import assert_true
-
 from pykalman import AdditiveUnscentedKalmanFilter, UnscentedKalmanFilter
 from pykalman.datasets import load_robot
 
@@ -56,38 +54,32 @@ def check_dims(n_dim_state, n_dim_obs, n_func_args, kf_cls, kwargs):
  kf._initialize_parameters()
  )
 
- assert_true(
+ assert (
  transition_functions.shape == (1,)
  if not 'transition_functions' in kwargs
  else (len(kwargs['transition_functions']),)
  )
- assert_true(
- all([len(inspect.getargspec(f).args) == n_func_args
+ assert all([len(inspect.getargspec(f).args) == n_func_args
  for f in transition_functions])
- )
- assert_true(transition_covariance.shape == (n_dim_state, n_dim_state))
- assert_true(
+ assert transition_covariance.shape == (n_dim_state, n_dim_state)
+ assert (
  observation_functions.shape == (1,)
  if not 'observation_functions' in kwargs
  else (len(kwargs['observation_functions']),)
  )
- assert_true(
- all([len(inspect.getargspec(f).args) == n_func_args
+ assert all([len(inspect.getargspec(f).args) == n_func_args
  for f in observation_functions])
- )
- assert_true(observation_covariance.shape == (n_dim_obs, n_dim_obs))
- assert_true(initial_state_mean.shape == (n_dim_state,))
- assert_true(
- initial_state_covariance.shape == (n_dim_state, n_dim_state)
- )
+ assert observation_covariance.shape == (n_dim_obs, n_dim_obs)
+ assert initial_state_mean.shape == (n_dim_state,)
+ assert initial_state_covariance.shape == (n_dim_state, n_dim_state)
 
 
 def test_unscented_sample():
  kf = build_unscented_filter(UnscentedKalmanFilter)
  (x, z) = kf.sample(100)
 
- assert_true(x.shape == (100, 2))
- assert_true(z.shape == (100, 1))
+ assert x.shape == (100, 2)
+ assert z.shape == (100, 1)
 
 
 def test_unscented_filter():
@@ -157,8 +149,8 @@ def test_additive_sample():
  kf = build_unscented_filter(AdditiveUnscentedKalmanFilter)
  (x, z) = kf.sample(100)
 
- assert_true(x.shape == (100, 2))
- assert_true(z.shape == (100, 1))
+ assert x.shape == (100, 2)
+ assert z.shape == (100, 1)
 
 
 def test_additive_filter():

pyproject.toml

@@ -29,7 +29,7 @@ dependencies = [
 [project.optional-dependencies]
 
 tests = [
- "nose"
+ "pytest"
 ]
 
 docs = [
@@ -38,7 +38,7 @@ docs = [
 ]
 
 all = [
- "nose",
+ "pytest",
  "Sphinx",
  "numpydoc"
 ]
@@ -52,7 +52,7 @@ Documentation = "https://pykalman.github.io/"
 file = "COPYING"
 
 [build-system]
-requires = ["setuptools", "wheel", "toml", "build"]
+requires = ["setuptools", "wheel", "build"]
 build-backend = "setuptools.build_meta"
 
 [tool.setuptools]