def __init__(self,
vectorizer='tfidf',
num_leaves=23,
learning_rate=0.01,
max_depth=-1,
num_boost_round=99999):
self.vectorizer = feature_extraction.get(vectorizer)
self.classifier = self.build_classifier(num_leaves, learning_rate,
max_depth, num_boost_round)
def __init__(self, question_encoder_shape=64, text_encoder_shape=64, learning_rate=0.001):
self.vectorizer = feature_extraction.get('word2vec')
self.num_features = config.WORD_VECTOR_DIM
self.question_encoder_shape = question_encoder_shape
self.text_encoder_shape = text_encoder_shape
# Tensorflow currently does not support Tensors with different lengths along a dimension.
self.batch_size = 1
self.learning_rate = learning_rate
self.classifier = self.build_classifier()
def __init__(self, vectorizer='tfidf', kernel='rbf', degree=3):
self.vectorizer = feature_extraction.get(vectorizer)
self.classifier = self.build_classifier(kernel, degree)
def __init__(self, vectorizer='tfidf'):
self.vectorizer = feature_extraction.get(vectorizer)
self.classifier = self.build_classifier()
def __init__(self, vectorizer='tfidf', n_neighbors=3):
self.vectorizer = feature_extraction.get(vectorizer)
self.classifier = self.build_classifier(n_neighbors)
def __init__(self, vectorizer='tfidf', max_depth=66, random_state=1):
self.vectorizer = feature_extraction.get(vectorizer)
self.classifier = self.build_classifier(max_depth, random_state)
def __init__(self, vectorizer='tfidf', random_state=None):
self.vectorizer = feature_extraction.get(vectorizer)
self.classifier = self.build_classifier(random_state)
def __init__(self, vectorizer='tfidf', max_depth=8, tree_method='auto'):
self.vectorizer = feature_extraction.get(vectorizer)
self.classifier = self.build_classifier(max_depth, tree_method)
def __init__(self, embedding_dim=64, batch_size=32):
self.vectorizer = feature_extraction.get('label_encoder')
self.embedding_dim = embedding_dim
self.batch_size = batch_size
self.classifier = self.build_classifier()