Merge pull request #774 from lilleswing/shuffle-test (c5d23c0e) · Commits · 钟慕尧 / deepchem

deepchem/data/datasets.py

+1 −1

Original line number	Diff line number	Diff line
		@@ -768,12 +768,12 @@ class DiskDataset(Dataset):
		y = np.reshape(y, (n_samples, -1))
		if w is not None:
		w = np.reshape(w, (n_samples, -1))
		n_tasks = y.shape[1]
		if ids is None:
		ids = np.arange(n_samples)
		if w is None:
		w = np.ones_like(y)
		if tasks is None:
		n_tasks = y.shape[1]
		tasks = np.arange(n_tasks)
		# raw_data = (X, y, w, ids)
		return DiskDataset.create_dataset(

deepchem/models/tensorflow_models/fcnet.py

+6 −3

Original line number	Diff line number	Diff line
		@@ -132,11 +132,12 @@ class TensorGraphMultiTaskClassifier(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		for epoch in range(epochs):
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):
		feed_dict = dict()
		if y_b is not None and not predict:
		@@ -292,11 +293,12 @@ class TensorGraphMultiTaskRegressor(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		for epoch in range(epochs):
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):
		feed_dict = dict()
		if y_b is not None and not predict:
		@@ -373,11 +375,12 @@ class TensorGraphMultiTaskFitTransformRegressor(TensorGraphMultiTaskRegressor):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		for epoch in range(epochs):
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):
		feed_dict = dict()
		if y_b is not None and not predict:

deepchem/models/tensorgraph/models/graph_models.py

+16 −9

Original line number	Diff line number	Diff line
		@@ -123,6 +123,7 @@ class WeaveTensorGraph(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		""" TensorGraph style implementation
		similar to deepchem.models.tf_new_models.graph_topology.AlternateWeaveTopology.batch_to_feed_dict
		@@ -132,7 +133,7 @@ class WeaveTensorGraph(TensorGraph):
		print('Starting epoch %i' % epoch)
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):

		feed_dict = dict()
		@@ -279,6 +280,7 @@ class DTNNTensorGraph(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		""" TensorGraph style implementation
		similar to deepchem.models.tf_new_models.graph_topology.DTNNGraphTopology.batch_to_feed_dict
		@@ -288,7 +290,7 @@ class DTNNTensorGraph(TensorGraph):
		print('Starting epoch %i' % epoch)
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):

		feed_dict = dict()
		@@ -430,6 +432,7 @@ class DAGTensorGraph(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		""" TensorGraph style implementation
		similar to deepchem.models.tf_new_models.graph_topology.DAGGraphTopology.batch_to_feed_dict
		@@ -439,7 +442,7 @@ class DAGTensorGraph(TensorGraph):
		print('Starting epoch %i' % epoch)
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):

		feed_dict = dict()
		@@ -575,13 +578,14 @@ class GraphConvTensorGraph(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		for epoch in range(epochs):
		if not predict:
		print('Starting epoch %i' % epoch)
		for ind, (X_b, y_b, w_b, ids_b) in enumerate(
		dataset.iterbatches(
		self.batch_size, pad_batches=True, deterministic=True)):
		self.batch_size, pad_batches=True, deterministic=deterministic)):
		d = {}
		for index, label in enumerate(self.my_labels):
		if self.mode == 'classification':
		@@ -704,6 +708,7 @@ class MPNNTensorGraph(TensorGraph):
		n_hidden=100,
		T=5,
		M=10,
		mode="regression",
		**kwargs):
		"""
		Parameters
		@@ -726,6 +731,7 @@ class MPNNTensorGraph(TensorGraph):
		self.n_hidden = n_hidden
		self.T = T
		self.M = M
		self.mode = mode
		super(MPNNTensorGraph, self).__init__(**kwargs)
		self.build_graph()

		@@ -789,6 +795,7 @@ class MPNNTensorGraph(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		""" Same generator as Weave models """
		for epoch in range(epochs):
		@@ -796,7 +803,7 @@ class MPNNTensorGraph(TensorGraph):
		print('Starting epoch %i' % epoch)
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):

		feed_dict = dict()

deepchem/models/tensorgraph/models/symmetry_function_regression.py

+2 −1

Original line number	Diff line number	Diff line
		@@ -167,13 +167,14 @@ class ANIRegression(TensorGraph):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		for epoch in range(epochs):
		if not predict:
		print('Starting epoch %i' % epoch)
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):

		feed_dict = dict()

deepchem/models/tensorgraph/tensor_graph.py

+14 −5

Original line number	Diff line number	Diff line
		@@ -65,8 +65,8 @@ class TensorGraph(Model):
		self.loss = None
		self.built = False
		self.queue_installed = False
		self.optimizer = Adam(
		learning_rate=learning_rate, beta1=0.9, beta2=0.999, epsilon=1e-7)
		self.optimizer = None
		self.learning_rate = learning_rate

		# Singular place to hold Tensor objects which don't serialize
		# These have to be reconstructed on restoring from pickle
		@@ -115,9 +115,11 @@ class TensorGraph(Model):
		dataset,
		nb_epoch=10,
		max_checkpoints_to_keep=5,
		checkpoint_interval=1000):
		checkpoint_interval=1000,
		deterministic=False):
		return self.fit_generator(
		self.default_generator(dataset, epochs=nb_epoch),
		self.default_generator(
		dataset, epochs=nb_epoch, deterministic=deterministic),
		max_checkpoints_to_keep, checkpoint_interval)

		def fit_generator(self,
		@@ -206,6 +208,7 @@ class TensorGraph(Model):
		dataset,
		epochs=1,
		predict=False,
		deterministic=True,
		pad_batches=True):
		if len(self.features) > 1:
		raise ValueError("More than one Feature, must use generator")
		@@ -216,7 +219,7 @@ class TensorGraph(Model):
		for epoch in range(epochs):
		for (X_b, y_b, w_b, ids_b) in dataset.iterbatches(
		batch_size=self.batch_size,
		deterministic=True,
		deterministic=deterministic,
		pad_batches=pad_batches):
		feed_dict = dict()
		if len(self.labels) == 1 and y_b is not None and not predict:
		@@ -587,6 +590,12 @@ class TensorGraph(Model):
		elif obj == "FileWriter":
		self.tensor_objects['FileWriter'] = tf.summary.FileWriter(self.model_dir)
		elif obj == 'Optimizer':
		if self.optimizer is None:
		self.optimizer = Adam(
		learning_rate=self.learning_rate,
		beta1=0.9,
		beta2=0.999,
		epsilon=1e-7)
		self.tensor_objects['Optimizer'] = self.optimizer._create_optimizer(
		self._get_tf('GlobalStep'))
		elif obj == 'train_op':

Admin message