def predict_distributed(model,
x=None,
batch_size=None,
verbose=0,
steps=None,
callbacks=None):
"""Predict loop for Distribution Strategies."""
distributed_training_utils.validate_inputs(x, None)
first_x_value = nest.flatten(x)[0]
if isinstance(first_x_value, np.ndarray):
steps, batch_size = distributed_training_utils.get_input_params(
model._distribution_strategy, first_x_value, steps,
batch_size, mode=ModeKeys.PREDICT)
batch_size = model._validate_or_infer_batch_size(batch_size, steps, x)
dataset = model._distribution_standardize_user_data(
x,
batch_size=batch_size,
allow_partial_batch=True)
if distributed_training_utils.is_tpu_strategy(model._distribution_strategy):
return experimental_tpu_predict_loop(
model, dataset, verbose=verbose, steps=steps, callbacks=callbacks)
else:
return training_arrays.predict_loop(
model,
dataset,
batch_size=batch_size,
verbose=verbose,
steps=steps,
callbacks=callbacks)
def evaluate_distributed(model,
x=None,
y=None,
batch_size=None,
verbose=1,
sample_weight=None,
steps=None,
callbacks=None):
"""Evaluate loop for Distribution Strategies."""
distributed_training_utils.validate_inputs(x, y,
model._distribution_strategy)
first_x_value = nest.flatten(x)[0]
if isinstance(first_x_value, np.ndarray):
steps, batch_size = distributed_training_utils.get_input_params(
model._distribution_strategy, first_x_value, steps, batch_size)
batch_size = model._validate_or_infer_batch_size(batch_size, steps, x)
dataset = model._distribution_standardize_user_data(
x, y, sample_weight=sample_weight, batch_size=batch_size)
if distributed_training_utils.is_tpu_strategy(
model._distribution_strategy):
return experimental_tpu_test_loop(model,
dataset,
verbose=verbose,
steps=steps,
callbacks=callbacks)
else:
return training_arrays.test_loop(model,
inputs=dataset,
batch_size=batch_size,
verbose=verbose,
steps=steps,
callbacks=callbacks)
def predict_distributed(model,
x=None,
batch_size=None,
verbose=0,
steps=None,
callbacks=None):
"""Predict loop for Distribution Strategies."""
distributed_training_utils.validate_inputs(
x, None, model._distribution_strategy, allow_partial_batch=True)
first_x_value = nest.flatten(x)[0]
if isinstance(first_x_value, np.ndarray):
steps, batch_size = distributed_training_utils.get_input_params(
model._distribution_strategy, first_x_value, steps,
batch_size, mode=ModeKeys.PREDICT)
batch_size = model._validate_or_infer_batch_size(batch_size, steps, x)
dataset = model._distribution_standardize_user_data(
x,
batch_size=batch_size,
allow_partial_batch=True)
if distributed_training_utils.is_tpu_strategy(model._distribution_strategy):
return experimental_tpu_predict_loop(
model, dataset, verbose=verbose, steps=steps, callbacks=callbacks)
else:
return training_arrays.predict_loop(
model,
dataset,
batch_size=batch_size,
verbose=verbose,
steps=steps,
callbacks=callbacks)
def evaluate_distributed(model,
x=None,
y=None,
batch_size=None,
verbose=1,
sample_weight=None,
steps=None,
callbacks=None):
"""Evaluate loop for Distribution Strategies."""
distributed_training_utils.validate_inputs(x, y, model._distribution_strategy)
first_x_value = nest.flatten(x)[0]
if isinstance(first_x_value, np.ndarray):
steps, batch_size = distributed_training_utils.get_input_params(
model._distribution_strategy, first_x_value, steps, batch_size)
batch_size = model._validate_or_infer_batch_size(batch_size, steps, x)
dataset = model._distribution_standardize_user_data(
x, y,
sample_weight=sample_weight,
batch_size=batch_size)
if distributed_training_utils.is_tpu_strategy(model._distribution_strategy):
return experimental_tpu_test_loop(
model, dataset, verbose=verbose, steps=steps, callbacks=callbacks)
else:
return training_arrays.test_loop(
model,
inputs=dataset,
batch_size=batch_size,
verbose=verbose,
steps=steps,
callbacks=callbacks)
def _process_batch_and_step_size(self, model, inputs, batch_size,
steps_per_epoch, mode):
first_x_value = nest.flatten(inputs)[0]
if isinstance(first_x_value, np.ndarray):
# Until support for partial batch is implemented across all
# functions and distribution strategy, we pass `mode` to selectively
# relax the constraint to consume all the training samples.
steps_per_epoch, batch_size = (dist_utils.get_input_params(
model._distribution_strategy,
first_x_value,
steps_per_epoch,
batch_size,
mode=mode))
return batch_size, steps_per_epoch
def fit_distributed(model,
x=None,
y=None,
batch_size=None,
epochs=1,
verbose=1,
callbacks=None,
validation_split=0.,
validation_data=None,
shuffle=True,
class_weight=None,
sample_weight=None,
initial_epoch=0,
steps_per_epoch=None,
validation_steps=None,
validation_freq=1):
"""Fit loop for Distribution Strategies."""
distributed_training_utils.validate_callbacks(callbacks, model.optimizer)
distributed_training_utils.validate_inputs(
x, y)
first_x_value = nest.flatten(x)[0]
if isinstance(first_x_value, np.ndarray):
# Until support for partial batch is implemented across all
# functions and distribution strategy, we pass `mode` to selectively
# relax the costraint to consume all the training samples.
steps_per_epoch, batch_size = (
distributed_training_utils.get_input_params(
model._distribution_strategy, first_x_value, steps_per_epoch,
batch_size, mode=ModeKeys.TRAIN))
batch_size = model._validate_or_infer_batch_size(
batch_size, steps_per_epoch, x)
dataset = model._distribution_standardize_user_data(
x, y,
sample_weight=sample_weight,
class_weight=class_weight,
batch_size=batch_size,
validation_split=validation_split,
shuffle=shuffle,
epochs=epochs)
if not distributed_training_utils.is_distributing_by_cloning(model):
with model._distribution_strategy.scope():
(dataset, _, _) = model._standardize_user_data(
dataset,
sample_weight=sample_weight,
class_weight=class_weight,
batch_size=batch_size,
validation_split=validation_split,
shuffle=shuffle)
val_dataset = None
if validation_data:
val_x, val_y, val_sample_weights = model._unpack_validation_data(
validation_data)
distributed_training_utils.validate_inputs(val_x, val_y)
first_valx_value = nest.flatten(val_x)[0]
if isinstance(first_valx_value, np.ndarray):
validation_steps, _ = distributed_training_utils.get_input_params(
model._distribution_strategy, first_valx_value, validation_steps,
batch_size, mode=ModeKeys.TEST)
val_dataset = model._distribution_standardize_user_data(
val_x, val_y,
sample_weight=val_sample_weights,
class_weight=None,
batch_size=batch_size,
validation_split=validation_split,
shuffle=shuffle,
allow_partial_batch=True)
elif validation_split:
raise ValueError('validation_split argument is not supported with '
'distribution strategies.')
if distributed_training_utils.is_tpu_strategy(model._distribution_strategy):
return experimental_tpu_fit_loop(
model,
dataset,
epochs=epochs,
verbose=verbose,
callbacks=callbacks,
val_dataset=val_dataset,
initial_epoch=initial_epoch,
steps_per_epoch=steps_per_epoch,
validation_steps=validation_steps,
validation_freq=validation_freq)
else:
return training_arrays.fit_loop(
model,
dataset,
batch_size=batch_size,
epochs=epochs,
verbose=verbose,
callbacks=callbacks,
val_inputs=val_dataset,
shuffle=shuffle,
initial_epoch=initial_epoch,
steps_per_epoch=steps_per_epoch,
validation_steps=validation_steps,
validation_freq=validation_freq,
steps_name='steps_per_epoch')
def fit_distributed(model,
x=None,
y=None,
batch_size=None,
epochs=1,
verbose=1,
callbacks=None,
validation_split=0.,
validation_data=None,
shuffle=True,
class_weight=None,
sample_weight=None,
initial_epoch=0,
steps_per_epoch=None,
validation_steps=None,
validation_freq=1):
"""Fit loop for Distribution Strategies."""
distributed_training_utils.validate_callbacks(callbacks, model.optimizer)
distributed_training_utils.validate_inputs(
x, y, model._distribution_strategy)
first_x_value = nest.flatten(x)[0]
if isinstance(first_x_value, np.ndarray):
# Until support for partial batch is implemented across all
# functions and distribution strategy, we pass `mode` to selectively
# relax the costraint to consume all the training samples.
steps_per_epoch, batch_size = (
distributed_training_utils.get_input_params(
model._distribution_strategy, first_x_value, steps_per_epoch,
batch_size, mode=ModeKeys.TRAIN))
batch_size = model._validate_or_infer_batch_size(
batch_size, steps_per_epoch, x)
dataset = model._distribution_standardize_user_data(
x, y,
sample_weight=sample_weight,
class_weight=class_weight,
batch_size=batch_size,
validation_split=validation_split,
shuffle=shuffle,
repeat=True)
val_dataset = None
if validation_data:
val_x, val_y, val_sample_weights = model._unpack_validation_data(
validation_data)
distributed_training_utils.validate_inputs(
val_x, val_y, model._distribution_strategy)
first_valx_value = nest.flatten(val_x)[0]
if isinstance(first_valx_value, np.ndarray):
validation_steps, _ = distributed_training_utils.get_input_params(
model._distribution_strategy, first_valx_value, validation_steps,
batch_size)
val_dataset = model._distribution_standardize_user_data(
val_x, val_y,
sample_weight=val_sample_weights,
class_weight=None,
batch_size=batch_size,
validation_split=validation_split,
shuffle=shuffle)
elif validation_split:
raise ValueError('validation_split argument is not supported with '
'distribution strategies.')
if distributed_training_utils.is_tpu_strategy(model._distribution_strategy):
return experimental_tpu_fit_loop(
model,
dataset,
epochs=epochs,
verbose=verbose,
callbacks=callbacks,
val_dataset=val_dataset,
initial_epoch=initial_epoch,
steps_per_epoch=steps_per_epoch,
validation_steps=validation_steps,
validation_freq=validation_freq)
else:
return training_arrays.fit_loop(
model,
dataset,
batch_size=batch_size,
epochs=epochs,
verbose=verbose,
callbacks=callbacks,
val_inputs=val_dataset,
shuffle=shuffle,
initial_epoch=initial_epoch,
steps_per_epoch=steps_per_epoch,
validation_steps=validation_steps,
validation_freq=validation_freq,
steps_name='steps_per_epoch')
