How to use the xenonpy.model.training.MSELoss function in xenonpy

To help you get started, we’ve selected a few xenonpy examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
def test_trainer_fit_1(data):
    model = deepcopy(data[0])
    trainer = Trainer(model=model, optimizer=Adam(), loss_func=MSELoss())
    trainer.fit(*data[1])
    assert trainer.total_iterations == 200
    assert trainer.total_epochs == 200

    trainer.fit(*data[1], epochs=20)
    assert trainer.total_iterations == 220
    assert trainer.total_epochs == 220

    trainer.reset()
    assert trainer.total_iterations == 0
    assert trainer.total_epochs == 0

    trainer.fit(*data[1], epochs=20)
    assert trainer.total_iterations == 20
    assert trainer.total_epochs == 20
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
def test_trainer_prediction_1(data):
    model = deepcopy(data[0])
    trainer = Trainer(model=model, optimizer=Adam(lr=0.1), loss_func=MSELoss(), epochs=200)
    trainer.extend(TensorConverter())
    trainer.fit(*data[1], *data[1])

    trainer = Trainer(model=model).extend(TensorConverter())
    y_p = trainer.predict(data[1][0])
    assert np.any(np.not_equal(y_p, data[1][1].numpy()))
    assert np.allclose(y_p, data[1][1].numpy(), rtol=0, atol=0.2)

    y_p, y_t = trainer.predict(*data[1])
    assert np.any(np.not_equal(y_p, y_t))
    assert np.allclose(y_p, y_t, rtol=0, atol=0.2)

    val_set = DataLoader(TensorDataset(*data[1]), batch_size=50)
    y_p, y_t = trainer.predict(dataset=val_set)
    assert np.any(np.not_equal(y_p, y_t))
    assert np.allclose(y_p, y_t, rtol=0, atol=0.2)
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
def test_trainer_fit_2(data):
    model = deepcopy(data[0])
    trainer = Trainer(model=model, optimizer=Adam(), loss_func=MSELoss(), epochs=20)
    trainer.fit(*data[1], *data[1])
    assert trainer.total_iterations == 20
    assert trainer.total_epochs == 20
    assert (trainer.x_val, trainer.y_val) == data[1]

    train_set = DataLoader(TensorDataset(*data[1]))
    val_set = DataLoader(TensorDataset(*data[1]))
    trainer.fit(training_dataset=train_set, validation_dataset=val_set)
    assert trainer.total_iterations == 2020
    assert trainer.total_epochs == 40
    assert isinstance(trainer.validate_dataset, DataLoader)
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
def test_trainer_fit_3(data):
    model = deepcopy(data[0])
    trainer = Trainer(model=model, optimizer=Adam(), loss_func=MSELoss(), epochs=5)
    trainer.fit(*data[1])
    assert len(trainer.checkpoints.keys()) == 0

    trainer.reset()
    assert trainer.total_iterations == 0
    assert trainer.total_epochs == 0
    assert len(trainer.get_checkpoint()) == 0

    trainer.fit(*data[1], checkpoint=True)
    assert len(trainer.get_checkpoint()) == 5
    assert isinstance(trainer.get_checkpoint(2), trainer.checkpoint_tuple)
    assert isinstance(trainer.get_checkpoint('cp_2'), trainer.checkpoint_tuple)

    with pytest.raises(TypeError, match='parameter  must be str or int'):
        trainer.get_checkpoint([])
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
def test_trainer_2(data):
    trainer = Trainer()
    with pytest.raises(RuntimeError, match='no model for training'):
        trainer.fit(*data[1])

    with pytest.raises(TypeError, match='parameter `m` must be a instance of '):
        trainer.model = {}

    trainer.model = data[0]
    assert isinstance(trainer.model, torch.nn.Module)
    with pytest.raises(RuntimeError, match='no loss function for training'):
        trainer.fit(*data[1])

    trainer.loss_func = MSELoss()
    assert trainer.loss_type == 'train_mse_loss'
    assert trainer.loss_func.__class__ == MSELoss
    with pytest.raises(RuntimeError, match='no optimizer for training'):
        trainer.fit(*data[1])

    trainer.optimizer = Adam()
    assert isinstance(trainer.optimizer, torch.optim.Adam)
    assert isinstance(trainer._optimizer_state, dict)
    assert isinstance(trainer._init_states, dict)

    trainer.lr_scheduler = ExponentialLR(gamma=0.99)
    assert isinstance(trainer.lr_scheduler, torch.optim.lr_scheduler.ExponentialLR)
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
def test_trainer_3(data):
    model = data[0]
    trainer = Trainer(model=model, optimizer=Adam(), loss_func=MSELoss())
    assert isinstance(trainer.model, torch.nn.Module)
    assert isinstance(trainer.optimizer, torch.optim.Adam)
    assert isinstance(trainer._optimizer_state, dict)
    assert isinstance(trainer._init_states, dict)
    assert trainer.clip_grad is None
    assert trainer.lr_scheduler is None

    trainer.lr_scheduler = ExponentialLR(gamma=0.1)
    assert isinstance(trainer.lr_scheduler, torch.optim.lr_scheduler.ExponentialLR)

    trainer.optimizer = SGD()
    assert isinstance(trainer.optimizer, torch.optim.SGD)

    trainer.clip_grad = ClipNorm(max_norm=0.4)
    assert isinstance(trainer.clip_grad, ClipNorm)
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
assert trainer.model is None
    assert trainer.optimizer is None
    assert trainer.lr_scheduler is None
    assert trainer.x_val is None
    assert trainer.y_val is None
    assert trainer.validate_dataset is None
    assert trainer._init_states is None
    assert trainer._optimizer_state is None
    assert trainer.total_epochs == 0
    assert trainer.total_iterations == 0
    assert trainer.training_info is None
    assert trainer.loss_type is None
    assert trainer.loss_func is None

    trainer = Trainer(optimizer=Adam(),
                      loss_func=MSELoss(),
                      lr_scheduler=ExponentialLR(gamma=0.99),
                      clip_grad=ClipValue(clip_value=0.1))
    assert isinstance(trainer._scheduler, ExponentialLR)
    assert isinstance(trainer._optim, Adam)
    assert isinstance(trainer.clip_grad, ClipValue)
    assert isinstance(trainer.loss_func, MSELoss)
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
assert isinstance(trainer.model, torch.nn.Module)
    assert isinstance(trainer._training_info, list)
    assert trainer.optimizer is None
    assert trainer.lr_scheduler is None
    assert trainer.x_val is None
    assert trainer.y_val is None
    assert trainer.validate_dataset is None
    assert trainer._optimizer_state is None
    assert trainer.total_epochs == 0
    assert trainer.total_iterations == 0
    assert trainer.loss_type is None
    assert trainer.loss_func is None

    trainer = Trainer.load(from_=checker.path,
                           optimizer=Adam(),
                           loss_func=MSELoss(),
                           lr_scheduler=ExponentialLR(gamma=0.99),
                           clip_grad=ClipValue(clip_value=0.1))
    assert isinstance(trainer._scheduler, ExponentialLR)
    assert isinstance(trainer._optim, Adam)
    assert isinstance(trainer.clip_grad, ClipValue)
    assert isinstance(trainer.loss_func, MSELoss)
github yoshida-lab / XenonPy / tests / models / test_trainer.py View on Github external
def test_trainer_fit_4(data):
    model = deepcopy(data[0])
    trainer = Trainer(model=model,
                      optimizer=Adam(),
                      loss_func=MSELoss(),
                      clip_grad=ClipValue(0.1),
                      lr_scheduler=ReduceLROnPlateau(),
                      epochs=10)

    count = 1
    for i in trainer(*data[1]):
        assert isinstance(i, dict)
        assert i['i_epoch'] == count
        if count == 3:
            trainer.early_stop('stop')
        count += 1

    assert trainer.total_epochs == 3
    assert trainer._early_stopping == (True, 'stop')

    trainer.reset()