Merge branch 'master' into fix-docs-build (c1f2ddf2) · Commits · 钟慕尧 / deepchem

contrib/rl/mcts.py

+5 −2

Original line number	Diff line number	Diff line
		@@ -5,7 +5,10 @@ from deepchem.models.optimizers import Adam
		from deepchem.models.tensorgraph.layers import Feature, Weights, Label, Layer
		import numpy as np
		import tensorflow as tf
		import collections
		try:
		from collections.abc import Sequence as SequenceCollection
		except:
		from collections import Sequence as SequenceCollection
		import copy
		import time

		@@ -109,7 +112,7 @@ class MCTS(object):
		self.n_search_episodes = n_search_episodes
		self.discount_factor = discount_factor
		self.value_weight = value_weight
		self._state_is_list = isinstance(env.state_shape[0], collections.Sequence)
		self._state_is_list = isinstance(env.state_shape[0], SequenceCollection)
		if optimizer is None:
		self._optimizer = Adam(learning_rate=0.001, beta1=0.9, beta2=0.999)
		else:

contrib/tensorflow_models/robust_multitask.py

+10 −5

Original line number	Diff line number	Diff line
		@@ -6,6 +6,11 @@ import warnings
		import numpy as np
		import tensorflow as tf

		try:
		from collections.abc import Sequence as SequenceCollection
		except:
		from collections import Sequence as SequenceCollection

		from deepchem.nn import model_ops

		class RobustMultitaskClassifier(MultiTaskClassifier):
		@@ -73,15 +78,15 @@ class RobustMultitaskClassifier(MultiTaskClassifier):

		n_layers = len(layer_sizes)
		assert n_layers == len(bypass_layer_sizes)
		if not isinstance(weight_init_stddevs, collections.Sequence):
		if not isinstance(weight_init_stddevs, SequenceCollection):
		weight_init_stddevs = [weight_init_stddevs] * n_layers
		if not isinstance(bypass_weight_init_stddevs, collections.Sequence):
		if not isinstance(bypass_weight_init_stddevs, SequenceCollection):
		bypass_weight_init_stddevs = [bypass_weight_init_stddevs] * n_layers
		if not isinstance(bias_init_consts, collections.Sequence):
		if not isinstance(bias_init_consts, SequenceCollection):
		bias_init_consts = [bias_init_consts] * n_layers
		if not isinstance(dropouts, collections.Sequence):
		if not isinstance(dropouts, SequenceCollection):
		dropouts = [dropouts] * n_layers
		if not isinstance(activation_fns, collections.Sequence):
		if not isinstance(activation_fns, SequenceCollection):
		activation_fns = [activation_fns] * n_layers

		# Add the input features.

deepchem/hyper/grid_search.py

+1 −1

Original line number	Diff line number	Diff line
		@@ -110,7 +110,7 @@ class GridHyperparamOpt(HyperparamOpt):
		hyperparams = params_dict.keys()
		hyperparam_vals = params_dict.values()
		for hyperparam_list in params_dict.values():
		assert isinstance(hyperparam_list, collections.Iterable)
		assert isinstance(hyperparam_list, collections.abc.Iterable)

		number_combinations = reduce(mul, [len(vals) for vals in hyperparam_vals])

deepchem/models/cnn.py

+9 −6

Original line number	Diff line number	Diff line
		@@ -6,7 +6,10 @@ from deepchem.models.layers import SwitchedDropout
		from deepchem.metrics import to_one_hot
		from tensorflow.keras.layers import Input, Dense, Reshape, Softmax, Dropout, Activation, Lambda
		import tensorflow.keras.layers as layers
		import collections
		try:
		from collections.abc import Sequence as SequenceCollection
		except:
		from collections import Sequence as SequenceCollection


		class CNN(KerasModel):
		@@ -128,15 +131,15 @@ class CNN(KerasModel):
		n_layers = len(layer_filters)
		if not isinstance(kernel_size, list):
		kernel_size = [kernel_size] * n_layers
		if not isinstance(strides, collections.Sequence):
		if not isinstance(strides, SequenceCollection):
		strides = [strides] * n_layers
		if not isinstance(weight_init_stddevs, collections.Sequence):
		if not isinstance(weight_init_stddevs, SequenceCollection):
		weight_init_stddevs = [weight_init_stddevs] * (n_layers + 1)
		if not isinstance(bias_init_consts, collections.Sequence):
		if not isinstance(bias_init_consts, SequenceCollection):
		bias_init_consts = [bias_init_consts] * (n_layers + 1)
		if not isinstance(dropouts, collections.Sequence):
		if not isinstance(dropouts, SequenceCollection):
		dropouts = [dropouts] * n_layers
		if not isinstance(activation_fns, collections.Sequence):
		if not isinstance(activation_fns, SequenceCollection):
		activation_fns = [activation_fns] * n_layers
		if weight_decay_penalty != 0.0:
		if weight_decay_penalty_type == 'l1':

deepchem/models/fcnet.py

+12 −9

Original line number	Diff line number	Diff line
		@@ -6,7 +6,10 @@ import time
		import numpy as np
		import tensorflow as tf
		import threading
		import collections
		try:
		from collections.abc import Sequence as SequenceCollection
		except:
		from collections import Sequence as SequenceCollection

		import deepchem as dc
		from deepchem.models import KerasModel
		@@ -94,13 +97,13 @@ class MultitaskClassifier(KerasModel):
		self.n_features = n_features
		self.n_classes = n_classes
		n_layers = len(layer_sizes)
		if not isinstance(weight_init_stddevs, collections.Sequence):
		if not isinstance(weight_init_stddevs, SequenceCollection):
		weight_init_stddevs = [weight_init_stddevs] * n_layers
		if not isinstance(bias_init_consts, collections.Sequence):
		if not isinstance(bias_init_consts, SequenceCollection):
		bias_init_consts = [bias_init_consts] * n_layers
		if not isinstance(dropouts, collections.Sequence):
		if not isinstance(dropouts, SequenceCollection):
		dropouts = [dropouts] * n_layers
		if not isinstance(activation_fns, collections.Sequence):
		if not isinstance(activation_fns, SequenceCollection):
		activation_fns = [activation_fns] * n_layers
		if weight_decay_penalty != 0.0:
		if weight_decay_penalty_type == 'l1':
		@@ -240,13 +243,13 @@ class MultitaskRegressor(KerasModel):
		self.n_tasks = n_tasks
		self.n_features = n_features
		n_layers = len(layer_sizes)
		if not isinstance(weight_init_stddevs, collections.Sequence):
		if not isinstance(weight_init_stddevs, SequenceCollection):
		weight_init_stddevs = [weight_init_stddevs] * (n_layers + 1)
		if not isinstance(bias_init_consts, collections.Sequence):
		if not isinstance(bias_init_consts, SequenceCollection):
		bias_init_consts = [bias_init_consts] * (n_layers + 1)
		if not isinstance(dropouts, collections.Sequence):
		if not isinstance(dropouts, SequenceCollection):
		dropouts = [dropouts] * n_layers
		if not isinstance(activation_fns, collections.Sequence):
		if not isinstance(activation_fns, SequenceCollection):
		activation_fns = [activation_fns] * n_layers
		if weight_decay_penalty != 0.0:
		if weight_decay_penalty_type == 'l1':

Admin message