def test_torch_shallow_neural_classifier_incremental(XOR):
X, y = XOR
model = torch_shallow_neural_classifier.TorchShallowNeuralClassifier(
hidden_dim=4, hidden_activation=nn.ReLU(), max_iter=100, eta=0.01)
model.fit(X, y, X_dev=X, dev_iter=1)
epochs = list(model.dev_predictions.keys())
assert epochs == list(range(1, 101))
assert all(len(v) == len(X) for v in model.dev_predictions.values())
def test_torch_shallow_neural_classifier(XOR):
"""Just makes sure that this code will run; it doesn't check that
it is creating good models.
"""
X, y = XOR
model = torch_shallow_neural_classifier.TorchShallowNeuralClassifier(
hidden_dim=4, hidden_activation=nn.ReLU(), max_iter=100, eta=0.01)
model.fit(X, y)
model.predict(X)
model.predict_proba(X)
def test_torch_shallow_neural_classifier_save_load(XOR):
X, y = XOR
mod = torch_shallow_neural_classifier.TorchShallowNeuralClassifier(
hidden_dim=4, hidden_activation=nn.ReLU(), max_iter=100, eta=0.01)
mod.fit(X, y)
mod.predict(X)
with tempfile.NamedTemporaryFile(mode='wb') as f:
name = f.name
mod.to_pickle(name)
mod2 = torch_shallow_neural_classifier.TorchShallowNeuralClassifier.from_pickle(
name)
mod2.predict(X)
mod2.fit(X, y)
}
],
[
np_shallow_neural_classifier.ShallowNeuralClassifier(
hidden_dim=5, max_iter=1, eta=1.0),
{
'hidden_dim': 10,
# Reset to ReLU:
'hidden_activation': lambda z: np.maximum(0, z),
'd_hidden_activation': lambda z: np.where(z > 0, 1, 0),
'max_iter': 10,
'eta': 0.1
}
],
[
torch_shallow_neural_classifier.TorchShallowNeuralClassifier(
hidden_dim=5, hidden_activation=nn.ReLU(), max_iter=1, eta=1.0),
{
'hidden_dim': 10,
'hidden_activation': nn.ReLU(),
'max_iter': 10,
'eta': 0.1
}
],
[
np_autoencoder.Autoencoder(hidden_dim=5, max_iter=1, eta=1.0),
{'hidden_dim': 10, 'max_iter': 10, 'eta': 0.1}
],
[
torch_autoencoder.TorchAutoencoder(
hidden_dim=5, hidden_activation=nn.ReLU(), max_iter=1, eta=1.0),
{
