Added maxlength (213ce517) · Commits · 钟慕尧 / deepchem

deepchem/data/data_loader.py

+10 −4

Original line number	Diff line number	Diff line
		@@ -875,7 +875,7 @@ class FASTALoader(DataLoader):
		learning tasks.
		"""

		def __init__(self, featurizer: Featurizer = OneHotFeaturizer, protein = False):
		def __init__(self, featurizer: Featurizer = OneHotFeaturizer, protein = False, max_length = 100):
		"""Initialize FASTALoader.

		Parameters
		@@ -886,6 +886,10 @@ class FASTALoader(DataLoader):
		protein: bool (default: False)
		Whether or not the sequence passed in is a protein sequence. If False,
		it is treated as a nucleic acid sequence.

		max_length: int (default: 100)
		The maximum length of a string in the FASTA file. If Featurizer = OneHotFeaturizer,
		all one hot encodings will be padded to this length.
		"""
		protein_charset = ('A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K',
		'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V',
		@@ -898,9 +902,11 @@ class FASTALoader(DataLoader):
		self.user_specified_features = featurizer.feature_fields
		if (featurizer == OneHotFeaturizer):
		if (protein):
		featurizer = OneHotFeaturizer(protein_charset)
		featurizer = OneHotFeaturizer(protein_charset, max_length)
		else:
		featurizer = OneHotFeaturizer(nucleic_charset, max_length)
		else:
		featurizer = OneHotFeaturizer(nucleic_charset)
		featurizer = featurizer()
		self.featurizer = featurizer

		def create_dataset(self,
		@@ -944,7 +950,7 @@ class FASTALoader(DataLoader):
		X = self.featurizer(sequences)
		logger.warning(f"TESTING FINAL FEATURIZED ARRAY: {X}")
		ids = np.ones(len(X))
		# (X, y, w, ids) TODO discuss shape
		# (X, y, w, ids)
		yield X, None, None, ids

		def _read_file(input_file: str, auto_add_annotations: bool=False):

Admin message