Cleanup (372d514e) · Commits · 钟慕尧 / deepchem

deepchem/models/init.py

+0 −2

Original line number	Diff line number	Diff line
		@@ -111,12 +111,10 @@ class Model(object):
		# TODO(rbharath/enf): We need a structured way to deal with potential GPU
		# memory overflows.
		batch_size = self.model_params["batch_size"]
		####################################################### DEBUG
		if "pad_batches" in self.model_params:
		pad_batches = self.model_params["pad_batches"]
		else:
		pad_batches = False
		####################################################### DEBUG
		for epoch in range(self.model_params["nb_epoch"]):
		log("Starting epoch %s" % str(epoch+1), self.verbosity)
		losses = []

deepchem/models/tensorflow_models/init.py

+0 −8

Original line number	Diff line number	Diff line
		@@ -142,12 +142,6 @@ class TensorflowGraph(object):
		task_str = str(task).zfill(len(str(self.num_tasks)))
		with self._shared_name_scope('cost_{}'.format(task_str)):
		with tf.name_scope('weighted'):
		####################################################### DEBUG
		#print("len(self.output)")
		#print(len(self.output))
		#print("self.output")
		#print(self.output)
		####################################################### DEBUG
		weighted_cost = self.cost(self.output[task], self.labels[task],
		self.weights[task])
		weighted_costs.append(weighted_cost)
		@@ -218,12 +212,10 @@ class TensorflowGraph(object):
		if shuffle:
		log("About to shuffle dataset before epoch start.", self.verbosity)
		dataset.shuffle()
		####################################################################### DEBUG
		for ind, (X_b, y_b, w_b, ids_b) in enumerate(
		dataset.iterbatches(batch_size, pad_batches=pad_batches)):
		if ind % log_every_N_batches == 0:
		log("On batch %d" % ind, self.verbosity)
		####################################################################### DEBUG
		# Run training op.
		feed_dict = self.construct_feed_dict(X_b, y_b, w_b, ids_b)
		fetches = self.output + [

deepchem/models/tensorflow_models/fcnet.py

+0 −8

Original line number	Diff line number	Diff line
		@@ -67,9 +67,7 @@ from deepchem.models.tensorflow_models import TensorflowClassifier
		from deepchem.models.tensorflow_models import TensorflowRegressor
		from deepchem.models.tensorflow_models import model_ops
		from deepchem.metrics import to_one_hot
		################################################### DEBUG
		from deepchem.datasets import pad_features
		################################################### DEBUG

		def softmax(x):
		"""Simple numpy softmax implementation
		@@ -336,13 +334,11 @@ class TensorflowMultiTaskRegressor(TensorflowRegressor):
		num_tasks = self.num_tasks
		outputs = []
		with self._get_shared_session().as_default():
		################################################### DEBUG
		n_samples = len(X)
		# Some tensorflow models can't handle variadic batches,
		# especially models using tf.pack, tf.split. Pad batch-size
		# to handle these cases.
		X = pad_features(self.model_params["batch_size"], X)
		################################################### DEBUG
		feed_dict = self.construct_feed_dict(X)
		data = self._get_shared_session().run(
		self.output, feed_dict=feed_dict)
		@@ -352,22 +348,18 @@ class TensorflowMultiTaskRegressor(TensorflowRegressor):
		batch_outputs = batch_outputs.transpose((1, 0, 2))
		elif batch_outputs.ndim == 2:
		batch_outputs = batch_outputs.transpose((1, 0))
		########################################################### DEBUG
		# Handle edge case when batch-size is 1.
		elif batch_outputs.ndim == 1:
		#print("X.shape, batch_outputs.shape")
		#print(X.shape, batch_outputs.shape)
		n_samples = len(X)
		batch_outputs = batch_outputs.reshape((n_samples, num_tasks))
		########################################################### DEBUG
		else:
		raise ValueError(
		'Unrecognized rank combination for output: %s' %
		(batch_outputs.shape))
		##################################################### DEBUG
		# Prune away any padding that was added
		batch_outputs = batch_outputs[:n_samples]
		##################################################### DEBUG
		outputs.append(batch_outputs)

		outputs = np.squeeze(np.concatenate(outputs))

Admin message