Cleanup of cruft code (864e1d1a) · Commits · 钟慕尧 / deepchem

deepchem/datasets/init.py

+0 −3

Original line number	Diff line number	Diff line
		@@ -397,9 +397,6 @@ def _df_to_numpy(df, feature_types, tasks):
		# perform common train/test split across all tasks
		n_samples = df.shape[0]
		n_tasks = len(tasks)
		#y = df[tasks].values
		print("df.keys()")
		print(df.keys())
		y = np.array([df[task].values for task in tasks])
		y = np.reshape(y, (n_samples, n_tasks))
		w = np.ones((n_samples, n_tasks))

+0 −9

Original line number	Diff line number	Diff line
		@@ -248,10 +248,8 @@ class TensorflowGraph(object):
		ValueError: If output and labels are not both 3D or both 2D.
		"""

		######### DEBUG
		if not self._restored_model:
		self.restore()
		######### DEBUG
		with self.graph.as_default():
		assert not model_ops.is_training()
		self.require_attributes(['output'])
		@@ -388,15 +386,10 @@ class TensorflowGraph(object):
		if self._restored_model:
		return
		with self.graph.as_default():
		print("RESTORING MODEL FROM CHECKPOINT")
		assert not model_ops.is_training()
		last_checkpoint = self._find_last_checkpoint()
		print("last_checkpoint")
		print(last_checkpoint)

		saver = tf.train.Saver()
		#saver.restore(self._get_shared_session(),
		# tf_utils.ParseCheckpoint(last_checkpoint))
		saver.restore(self._get_shared_session(),
		last_checkpoint)
		self._restored_model = True
		@@ -574,14 +567,12 @@ class TensorflowModel(Model):
		Makes predictions on batch of data.
		"""
		return self.eval_model.predict_on_batch(X)
		#return self.train_model.predict_on_batch(X)

		def predict_proba_on_batch(self, X):
		"""
		Makes predictions on batch of data.
		"""
		return self.eval_model.predict_proba_on_batch(X)
		#return self.train_model.predict_proba_on_batch(X)

		def save(self):
		"""

+1 −10

Original line number	Diff line number	Diff line
		@@ -133,7 +133,6 @@ class TensorflowMultiTaskClassifier(TensorflowClassifier):
		bias_init=tf.constant(value=bias_init_consts[i],
		shape=[layer_sizes[i]])))
		layer = model_ops.Dropout(layer, dropouts[i])
		#layer = tf.nn.dropout(layer, dropouts[i])
		prev_layer = layer
		prev_layer_size = layer_sizes[i]

		@@ -186,14 +185,10 @@ class TensorflowMultiTaskClassifier(TensorflowClassifier):
		AssertionError: If model is not in evaluation mode.
		ValueError: If output and labels are not both 3D or both 2D.
		"""
		######### DEBUG
		if not self._restored_model:
		self.restore()
		######### DEBUG
		with self.graph.as_default():
		########## DEBUG
		assert not model_ops.is_training()
		########## DEBUG
		self.require_attributes(['output'])

		# run eval data through the model
		@@ -330,14 +325,10 @@ class TensorflowMultiTaskRegressor(TensorflowRegressor):
		AssertionError: If model is not in evaluation mode.
		ValueError: If output and labels are not both 3D or both 2D.
		"""
		########## DEBUG
		if not self._restored_model:
		self.restore()
		########## DEBUG
		with self.graph.as_default():
		########### DEBUG
		#assert not model_ops.is_training()
		########### DEBUG
		assert not model_ops.is_training()
		self.require_attributes(['output'])

		# run eval data through the model

+0 −18

Original line number	Diff line number	Diff line
		@@ -68,19 +68,6 @@ class Evaluator(object):
		mode = metrics[0].mode
		if mode == "classification":
		y_pred = self.model.predict_proba(self.dataset, self.transformers)
		############ DEBUG
		##import sklearn
		##from deepchem.metrics import to_one_hot
		##print("compute_model_performance()")
		##print("y_pred.shape, y.shape")
		##print(y_pred.shape, y.shape)
		##print("sklearn.metrics.log_loss(to_one_hot(y), np.squeeze(y_pred))")
		##print(sklearn.metrics.log_loss(to_one_hot(y), np.squeeze(y_pred)))
		##print("sklearn.metrics.roc_auc_score(to_one_hot(y), np.squeeze(y_pred))")
		##print(sklearn.metrics.roc_auc_score(to_one_hot(y), np.squeeze(y_pred)))
		##print("sklearn.metrics.roc_auc_score(to_one_hot(y), np.squeeze(y_pred), sample_weight=w)")
		##print(sklearn.metrics.roc_auc_score(to_one_hot(y), np.squeeze(y_pred), sample_weight=w))
		############ DEBUG

		else:
		y_pred = self.model.predict(self.dataset, self.transformers)
		@@ -94,9 +81,4 @@ class Evaluator(object):
		for metric in metrics:
		multitask_scores[metric.name] = metric.compute_metric(y, y_pred, w)

		########## DEBUG
		print("multitask_scores")
		print(multitask_scores)
		########## DEBUG

		return multitask_scores