finishing (6f2b005d) · Commits · 钟慕尧 / deepchem

deepchem/models/tensorflow_models/IRV.py

+36 −16

Original line number	Diff line number	Diff line
		@@ -79,6 +79,28 @@ class IRVLayer(Layer):
		def set_tensors(self, tensor):
		self.V, self.W, self.b, self.b2, self.out_tensor, self.trainable_weights, self.variables = tensor

		class IRVRegularize(Layer):
		""" This Layer extracts the trainable weights in IRVLayer
		and return the their L2-norm
		"""
		def __init__(self,
		IRVLayer,
		penalty=0.0,
		**kwargs):
		self.IRVLayer = IRVLayer
		self.penalty = penalty
		super(IRVRegularize, self).__init__(**kwargs)

		def create_tensor(self, in_layers=None, set_tensors=True, **kwargs):
		assert self.IRVLayer.out_tensor is not None, "IRVLayer must be built first"
		out_tensor = tf.nn.l2_loss(self.IRVLayer.W) + \
		tf.nn.l2_loss(self.IRVLayer.V) + tf.nn.l2_loss(self.IRVLayer.b) + \
		tf.nn.l2_loss(self.IRVLayer.b2)
		out_tensor = out_tensor * self.penalty
		if set_tensors:
		self.out_tensor = out_tensor
		return out_tensor

		class Slice(Layer):
		""" Choose a slice of input given axis and order
		"""
		@@ -118,9 +140,7 @@ class TensorflowMultiTaskIRVClassifier(TensorGraph):
		def __init__(self,
		n_tasks,
		K=10,
		n_classes=2,
		penalty=0.0,
		penalty_type="l2",
		mode="classification",
		**kwargs):
		"""Initialize TensorflowMultiTaskIRVClassifier
		@@ -131,12 +151,8 @@ class TensorflowMultiTaskIRVClassifier(TensorGraph):
		Number of tasks
		K: int
		Number of nearest neighbours used in classification
		n_classes: int
		number of different labels
		penalty: float
		Amount of penalty (l2 or l1 applied)
		penalty_type: str
		Either "l2" or "l1"

		"""
		self.n_tasks = n_tasks
		@@ -144,7 +160,6 @@ class TensorflowMultiTaskIRVClassifier(TensorGraph):
		self.n_features = 2 * self.K * self.n_tasks
		print("n_features after fit_transform: %d" % int(self.n_features))
		self.penalty = penalty
		self.penalty_type = penalty_type
		super(TensorflowMultiTaskIRVClassifier, self).__init__(**kwargs)
		self.build_graph()

		@@ -169,7 +184,8 @@ class TensorflowMultiTaskIRVClassifier(TensorGraph):
		costs.append(cost)
		all_cost = Concat(in_layers=costs, axis=1)
		self.weights = Weights(shape=(None, self.n_tasks))
		loss = WeightedError(in_layers=[all_cost, self.weights])
		loss = WeightedError(in_layers=[all_cost, self.weights]) + \
		IRVRegularize(predictions, self.penalty, in_layers=[predictions])
		self.set_loss(loss)

		def default_generator(self,
		@@ -197,12 +213,16 @@ class TensorflowMultiTaskIRVClassifier(TensorGraph):

		yield feed_dict

		def predict_on_generator(self, generator, transformers=[], outputs=None):
		out = super(TensorflowMultiTaskIRVClassifier, self).predict_on_generator(
		generator, transformers=[], outputs=outputs)
		if outputs is None:
		outputs = self.outputs
		if len(outputs) > 1:
		def predict(self, dataset, transformers=[], outputs=None):
		out = super(TensorflowMultiTaskIRVClassifier, self).predict(
		dataset, transformers=transformers, outputs=outputs)
		out = np.concatenate(out, axis=1)
		out = np.round(out).astype(int)
		return out

		def predict_proba(self, dataset, transformers=[], outputs=None):
		out = super(TensorflowMultiTaskIRVClassifier, self).predict_proba(
		dataset, transformers=transformers, outputs=outputs)
		out = np.concatenate(out, axis=1)
		out = undo_transforms(out, transformers)
		out = np.stack([1-out, out], axis=2)
		return out
		No newline at end of file

deepchem/models/tests/test_overfit.py

+0 −1

Original line number	Diff line number	Diff line
		@@ -627,7 +627,6 @@ class TestOverfit(test_util.TensorFlowTestCase):

		# Fit trained model
		model.fit(dataset_trans)
		model.save()

		# Eval model on train
		scores = model.evaluate(dataset_trans, [classification_metric])

deepchem/molnet/preset_hyper_parameters.py

+0 −1

Original line number	Diff line number	Diff line
		@@ -41,7 +41,6 @@ hps['logreg'] = {
		}
		hps['irv'] = {
		'penalty': 0.,
		'penalty_type': 'l2',
		'batch_size': 50,
		'nb_epoch': 10,
		'learning_rate': 0.001,

deepchem/molnet/run_benchmark_models.py

+2 −3

Original line number	Diff line number	Diff line
		@@ -161,7 +161,6 @@ def benchmark_classification(train_dataset,

		elif model_name == 'irv':
		penalty = hyper_parameters['penalty']
		penalty_type = hyper_parameters['penalty_type']
		batch_size = hyper_parameters['batch_size']
		nb_epoch = hyper_parameters['nb_epoch']
		learning_rate = hyper_parameters['learning_rate']
		@@ -179,10 +178,10 @@ def benchmark_classification(train_dataset,
		len(tasks),
		K=n_K,
		penalty=penalty,
		penalty_type=penalty_type,
		batch_size=batch_size,
		learning_rate=learning_rate,
		seed=seed)
		random_seed=seed,
		mode='classification')

		elif model_name == 'graphconv':
		batch_size = hyper_parameters['batch_size']

examples/hiv/hiv_irv.py

+0 −1

Original line number	Diff line number	Diff line
		@@ -29,7 +29,6 @@ model = dc.models.TensorflowMultiTaskIRVClassifier(

		# Fit trained model
		model.fit(train_dataset)
		model.save()

		print("Evaluating model")
		train_scores = model.evaluate(train_dataset, [metric], transformers)

Admin message