def _make_new_term_doc_matrix(self, new_X, new_mX, new_y, new_term_idx_store, new_category_idx_store, new_metadata_idx_store, new_y_mask): if self._parsed_col is not None and self._parsed_col in self._df: return ParsedCorpus(self._df[new_y_mask], new_X, new_mX, new_y, new_term_idx_store, new_category_idx_store, new_metadata_idx_store, self._parsed_col, self._category_col) else: return CorpusDF(self._df[new_y_mask], new_X, new_mX, new_y, self._text_col, new_term_idx_store, new_category_idx_store, new_metadata_idx_store, self._df[self._text_col][new_y_mask])
def _make_new_term_doc_matrix(self, new_X, new_mX, new_y,
new_term_idx_store, new_category_idx_store,
new_metadata_idx_store, new_y_mask):
return ParsedCorpus(self._df[new_y_mask], new_X, new_mX, new_y,
new_term_idx_store, new_category_idx_store,
new_metadata_idx_store, self._parsed_col,
self._category_col)
def build(self): '''Constructs the term doc matrix. Returns ------- scattertext.ParsedCorpus.ParsedCorpus ''' self._y = self._get_y_and_populate_category_idx_store() self._df.apply(self._add_to_x_factory, axis=1) self._X = self._X_factory.set_last_row_idx(len(self._y) - 1).get_csr_matrix() self._mX = self._mX_factory.set_last_row_idx(len(self._y) - 1).get_csr_matrix() if self._parsed_col is not None and self._parsed_col in self._df: return ParsedCorpus(self._df, self._X, self._mX, self._y, self._term_idx_store, self._category_idx_store, self._metadata_idx_store, self._parsed_col, self._category_col) else: return CorpusDF(self._df, self._X, self._mX, self._y, self._text_col, self._term_idx_store, self._category_idx_store, self._metadata_idx_store)
def build(self, show_progress=False):
'''Constructs the term doc matrix.
Returns
-------
scattertext.ParsedCorpus.ParsedCorpus
'''
y = self._get_y_and_populate_category_idx_store(
self._df[self._category_col])
if show_progress is True:
self._df.progress_apply(self._add_to_x_factory, axis=1)
else:
self._df.apply(self._add_to_x_factory, axis=1)
self._mX = self._mX_factory.set_last_row_idx(len(y) -
1).get_csr_matrix()
return ParsedCorpus(
df=self._df,
X=self._X_factory.set_last_row_idx(len(y) - 1).get_csr_matrix(),
mX=self._mX_factory.set_last_row_idx(len(y) - 1).get_csr_matrix(),
y=y,
term_idx_store=self._term_idx_store,
category_idx_store=self._category_idx_store,
metadata_idx_store=self._metadata_idx_store,
parsed_col=self._parsed_col,
category_col=self._category_col)
def build(self):
'''Constructs the term doc matrix.
Returns
-------
scattertext.ParsedCorpus.ParsedCorpus
'''
self._y = self._get_y_and_populate_category_idx_store()
self._df.apply(self._add_to_x_factory, axis=1)
self._X = self._X_factory.get_csr_matrix()
self._mX = self._mX_factory.get_csr_matrix()
return ParsedCorpus(self._df, self._X, self._mX, self._y,
self._term_idx_store, self._category_idx_store,
self._metadata_idx_store, self._parsed_col,
self._category_col)
def _make_new_term_doc_matrix(self, new_X=None, new_mX=None, new_y=None, new_term_idx_store=None, new_category_idx_store=None, new_metadata_idx_store=None, new_y_mask=None): return ParsedCorpus(self._df[new_y_mask] if new_y_mask else self._df, self._X if new_X is None else new_X, self._mX if new_mX is None else new_mX, self._y if new_y is None else new_y, self._term_idx_store if new_term_idx_store is None else new_term_idx_store, self._category_idx_store if new_category_idx_store is None else new_category_idx_store, self._metadata_idx_store if new_metadata_idx_store is None else new_metadata_idx_store, self._parsed_col, self._category_col)
def build(self):
'''
Returns
-------
CorpusDF
'''
if self.text_df is not None:
if self.parsed_col is not None:
if self.category_col is None:
self.text_df = self.text_df.assign(
Category=self.category_idx_store.getvalbatch(self.y))
self.category_col = 'Category'
return ParsedCorpus(
df=self.text_df,
X=self.X,
mX=self.mX,
y=self.y,
parsed_col=self.parsed_col,
term_idx_store=self.term_idx_store,
category_idx_store=self.category_idx_store,
metadata_idx_store=self.metadata_idx_store,
unigram_frequency_path=self.unigram_frequency_path,
category_col=self.category_col)
elif self.text_col is not None:
return CorpusDF(
df=self.text_df,
X=self.X,
mX=self.mX,
y=self.y,
text_col=self.text_col,
term_idx_store=self.term_idx_store,
category_idx_store=self.category_idx_store,
metadata_idx_store=self.metadata_idx_store,
unigram_frequency_path=self.unigram_frequency_path)
return TermDocMatrix(
X=self.X,
mX=self.mX,
y=self.y,
term_idx_store=self.term_idx_store,
category_idx_store=self.category_idx_store,
metadata_idx_store=self.metadata_idx_store,
unigram_frequency_path=self.unigram_frequency_path)