def _fit_predictive_model(self, seqs, neg_seqs=None):
# duplicate iterator
pos_seqs, pos_seqs_ = tee(seqs)
pos_graphs = mp_pre_process(pos_seqs,
pre_processor=sequence_to_eden,
n_blocks=self.pre_processor_n_blocks,
block_size=self.pre_processor_block_size,
n_jobs=self.pre_processor_n_jobs)
if neg_seqs is None:
# shuffle seqs to obtain negatives
neg_seqs = seq_to_seq(pos_seqs_,
modifier=shuffle_modifier,
times=self.negative_ratio,
order=self.shuffle_order)
neg_graphs = mp_pre_process(neg_seqs,
pre_processor=sequence_to_eden,
n_blocks=self.pre_processor_n_blocks,
block_size=self.pre_processor_block_size,
n_jobs=self.pre_processor_n_jobs)
# fit discriminative estimator
self.estimator = fit(pos_graphs,
neg_graphs,
vectorizer=self.vectorizer,
n_iter_search=self.n_iter_search,
n_jobs=self.n_jobs,
n_blocks=self.n_blocks,
block_size=self.block_size,
random_state=self.random_state)
def _fit_predictive_model(self, seqs, neg_seqs=None):
# duplicate iterator
pos_seqs, pos_seqs_ = tee(seqs)
pos_graphs = mp_pre_process(pos_seqs, pre_processor=sequence_to_eden,
n_blocks=self.pre_processor_n_blocks,
block_size=self.pre_processor_block_size,
n_jobs=self.pre_processor_n_jobs)
if neg_seqs is None:
# shuffle seqs to obtain negatives
neg_seqs = seq_to_seq(pos_seqs_,
modifier=shuffle_modifier,
times=self.negative_ratio,
order=self.shuffle_order)
neg_graphs = mp_pre_process(neg_seqs, pre_processor=sequence_to_eden,
n_blocks=self.pre_processor_n_blocks,
block_size=self.pre_processor_block_size,
n_jobs=self.pre_processor_n_jobs)
# fit discriminative estimator
self.estimator = fit(pos_graphs, neg_graphs,
vectorizer=self.vectorizer,
n_iter_search=self.n_iter_search,
n_jobs=self.n_jobs,
n_blocks=self.n_blocks,
block_size=self.block_size,
random_state=self.random_state)
def annotate(self, iterable):
assert(is_iterable(iterable)), 'Not iterable'
graphs = mp_pre_process(iterable,
pre_processor=self.pre_processor,
pre_processor_args=self.pre_processor_args,
n_blocks=self.pre_processor_n_blocks,
block_size=self.pre_processor_block_size,
n_jobs=self.pre_processor_n_jobs)
return self.vectorizer.annotate(graphs, self.estimator)
def annotate(self, iterable):
assert (is_iterable(iterable)), 'Not iterable'
graphs = mp_pre_process(iterable,
pre_processor=self.pre_processor,
pre_processor_args=self.pre_processor_args,
n_blocks=self.pre_processor_n_blocks,
block_size=self.pre_processor_block_size,
n_jobs=self.pre_processor_n_jobs)
return self.vectorizer.annotate(graphs, self.estimator)
def _data_matrix(self, iterable, fit_vectorizer=False):
assert(is_iterable(iterable)), 'Not iterable'
graphs = mp_pre_process(iterable,
pre_processor=self.pre_processor,
pre_processor_args=self.pre_processor_args,
n_blocks=self.pre_processor_n_blocks,
block_size=self.pre_processor_block_size,
n_jobs=self.pre_processor_n_jobs)
graphs, graphs_ = tee(graphs)
self.vectorizer.set_params(**self.vectorizer_args)
if fit_vectorizer:
self.vectorizer.fit(graphs_)
X = vectorize(graphs, vectorizer=self.vectorizer, n_jobs=self.n_jobs, n_blocks=self.n_blocks)
return X
def seq_to_data_matrix(self, sequences=None):
# Transform sequences to matrix
graphs = mp_pre_process(sequences, pre_processor=self.pre_processor, pre_processor_args={}, n_jobs=-1)
seq_data_matrix = vectorize(graphs, vectorizer=self.vectorizer, n_jobs=-1)
# Densify the matrix
seq_data_matrx = seq_data_matrix.toarray()
# Standardize the matrix
self.scale.fit(seq_data_matrx)
std_seq_data_matrx = self.scale.transform(seq_data_matrx)
return std_seq_data_matrx
