Starting on digesting keras code (47057e8e) · Commits · 钟慕尧 / deepchem

deepchem/nn/activations.py

+10 −10

Original line number	Diff line number	Diff line
		#from __future__ import absolute_import
		import six
		from keras import backend as K

		def get_from_module(identifier, module_params, module_name,
		@@ -24,8 +25,7 @@ def get_from_module(identifier, module_params, module_name,
		# Raises
		ValueError: if the identifier cannot be found.
		"""
		#if isinstance(identifier, str) or isinstance(identifier, unicode):
		assert not isinstance(identifier, dict)
		if isinstance(identifier, six.string_types):
		res = module_params.get(identifier)
		if not res:
		raise ValueError('Invalid ' + str(module_name) + ': ' +
		@@ -36,14 +36,14 @@ def get_from_module(identifier, module_params, module_name,
		return res(**kwargs)
		else:
		return res
		#elif isinstance(identifier, dict):
		# name = identifier.pop('name')
		# res = module_params.get(name)
		# if res:
		# return res(**identifier)
		# else:
		# raise ValueError('Invalid ' + str(module_name) + ': ' +
		# str(identifier))
		elif isinstance(identifier, dict):
		name = identifier.pop('name')
		res = module_params.get(name)
		if res:
		return res(**identifier)
		else:
		raise ValueError('Invalid ' + str(module_name) + ': ' +
		str(identifier))
		return identifier

		def softmax(x):

deepchem/nn/copy.py

+21 −30

Original line number	Diff line number	Diff line
		"""
		Copies Classes from keras to remove dependency.

		Most of this code is copied over from Keras. Hoping to use as a staging
		area while we remove our Keras dependency.
		"""
		from __future__ import print_function
		from __future__ import division
		from __future__ import unicode_literals

		__author__ = "Bharath Ramsundar"
		__copyright__ = "Copyright 2016, Stanford University"
		__license__ = "GPL"

		from . import initializations
		from . import regularizers
		from . import activations
		@@ -108,21 +107,30 @@ class Node(object):
		# and for each layer, which node and which
		# tensor output of each node.

		self.inbound_layers = inbound_layers # List of layer instances
		self.node_indices = node_indices # List of integers, 1:1 mapping with inbound_layers.
		self.tensor_indices = tensor_indices # List of integers, 1:1 mapping with inbound_layers.
		# List of layer instances
		self.inbound_layers = inbound_layers
		# List of integers, 1:1 mapping with inbound_layers.
		self.node_indices = node_indices
		# List of integers, 1:1 mapping with inbound_layers.
		self.tensor_indices = tensor_indices

		# Tensor inputs and outputs of outbound_layer.
		self.input_tensors = input_tensors # List of tensors. 1:1 mapping with inbound_layers.
		self.output_tensors = output_tensors # List of tensors, created by outbound_layer.call().
		# List of tensors. 1:1 mapping with inbound_layers.
		self.input_tensors = input_tensors
		# List of tensors, created by outbound_layer.call().
		self.output_tensors = output_tensors

		# input and output masks
		self.input_masks = input_masks # List of tensors, 1:1 mapping with input_tensor.
		self.output_masks = output_masks # List of tensors, created by outbound_layer.compute_mask().
		# List of tensors, 1:1 mapping with input_tensor.
		self.input_masks = input_masks
		# List of tensors, created by outbound_layer.compute_mask().
		self.output_masks = output_masks

		# input and output shapes
		self.input_shapes = input_shapes # List of shape tuples, shapes of input_tensors.
		self.output_shapes = output_shapes # List of shape tuples, shapes of output_tensors.
		# List of shape tuples, shapes of input_tensors.
		self.input_shapes = input_shapes
		# List of shape tuples, shapes of output_tensors.
		self.output_shapes = output_shapes

		# Add nodes to all layers involved.
		for layer in inbound_layers:
		@@ -204,8 +212,6 @@ class Node(object):
		'node_indices': self.node_indices,
		'tensor_indices': self.tensor_indices}



		class Layer(object):
		"""Abstract base layer class.

		@@ -357,21 +363,6 @@ class Layer(object):
		def non_trainable_weights(self, weights):
		self._non_trainable_weights = weights

		@property
		def regularizers(self):
		warnings.warn('The `regularizers` property of '
		'layers/models is deprecated. '
		'Regularization losses are now managed via the `losses` '
		'layer/model property.')
		return []

		@regularizers.setter
		def regularizers(self, _):
		warnings.warn('The `regularizers` property of layers/models '
		'is deprecated. '
		'Regularization losses are now managed via the `losses` '
		'layer/model property.')

		def create_input_layer(self, batch_input_shape,
		input_dtype=None, name=None):
		if not name:

Admin message