Cleanup (eb301f35) · Commits · 钟慕尧 / deepchem

deepchem/datasets/init.py

+2 −11

Original line number	Diff line number	Diff line
		@@ -183,11 +183,6 @@ class Dataset(object):
		Returns minibatches from dataset.
		"""
		for i, (X, y, w, ids) in enumerate(self.itershards()):
		######################## DEBUG
		print("Dataset.iterbatches()")
		print("i, X.shape, y.shape")
		print(i, X.shape, y.shape)
		######################## DEBUG
		nb_sample = np.shape(X)[0]
		if batch_size is None:
		shard_batch_size = nb_sample
		@@ -403,17 +398,11 @@ def _df_to_numpy(df, feature_types, tasks):
		feature_list.append(datapoint[feature_type])
		try:
		features = np.squeeze(np.concatenate(feature_list))
		################################# DEBUG
		#print("features.size: %s" % str(features.size))
		#print("features.size == 0")
		#print(features.size == 0)
		if features.size == 0:
		#print("CONTINUING")
		features = np.zeros(feature_shape)
		tensors.append(features)
		missing[ind, :] = 1
		continue
		################################# DEBUG
		for feature_ind, val in enumerate(features):
		if features[feature_ind] == "":
		features[feature_ind] = 0.
		@@ -441,6 +430,8 @@ def _df_to_numpy(df, feature_types, tasks):
		y[ind, task] = 0.
		w[ind, task] = 0.

		# Adding this assertion in to avoid ill-formed outputs.
		assert len(sorted_ids) == len(x) == len(y) == len(w)
		return sorted_ids, x.astype(float), y.astype(float), w.astype(float)

		def compute_mean_and_std(df):

+0 −3

Original line number	Diff line number	Diff line
		@@ -24,7 +24,6 @@ class TestDrop(TestAPI):
		verbosity = "high"
		len_full = 25

		## This is for good debug (to make sure nasty state isn't being passed around)
		current_dir = os.path.dirname(os.path.realpath(__file__))
		feature_dir = os.path.join(self.base_dir, "features")
		samples_dir = os.path.join(self.base_dir, "samples")
		@@ -54,8 +53,6 @@ class TestDrop(TestAPI):
		dataset = Dataset(data_dir=full_dir, samples=featurized_samples,
		featurizers=featurizers, tasks=emols_tasks,
		verbosity=verbosity, reload=reload)
		print("len(dataset)")
		print(len(dataset))

		X, y, w, ids = dataset.to_numpy()
		print("ids.shape, X.shape, y.shape, w.shape")

+0 −8

Original line number	Diff line number	Diff line
		@@ -257,14 +257,6 @@ class DataFeaturizer(object):
		if ind % self.log_every_n == 0:
		log("Featurizing sample %d" % ind, self.verbosity)
		mol = Chem.MolFromSmiles(smiles)
		################################ DEBUG
		if mol is None:
		#print("RDKit loading failed at %s" % smiles)
		output = featurizer.featurize([mol], verbosity=self.verbosity)
		#print("featurizer returned %s" % str(output))
		#print("type(output)")
		#print(type(output))
		################################ DEBUG
		features.append(featurizer.featurize([mol], verbosity=self.verbosity))
		else:
		def featurize_wrapper(smiles, dilled_featurizer):

+0 −14

Original line number	Diff line number	Diff line
		@@ -137,29 +137,15 @@ class Model(object):
		"""
		y_preds = []
		batch_size = self.model_params["batch_size"]
		######################## DEBUG
		print("Model.predict()")
		print("len(dataset)")
		print(len(dataset))
		######################## DEBUG
		for (X_batch, y_batch, w_batch, ids_batch) in dataset.iterbatches(batch_size):
		y_pred_batch = np.reshape(self.predict_on_batch(X_batch), y_batch.shape)
		y_pred_batch = undo_transforms(y_pred_batch, transformers)
		y_preds.append(y_pred_batch)
		######################## DEBUG
		print("y_batch.shape, y_pred_batch.shape")
		print(y_batch.shape, y_pred_batch.shape)
		######################## DEBUG
		y_pred = np.vstack(y_preds)

		# The iterbatches does padding with zero-weight examples on the last batch.
		# Remove padded examples.
		n_samples, n_tasks = len(dataset), len(self.tasks)
		######################## DEBUG
		print("Model.predict()")
		print("y_pred.shape, n_samples, n_tasks")
		print(y_pred.shape, n_samples, n_tasks)
		######################## DEBUG
		y_pred = np.reshape(y_pred, (n_samples, n_tasks))
		return y_pred