def repetitive_group_recognition(self, show=False, clean_attrs=True):
df_nontext = rep.recog_repetition_nontext(self, show)
df_text = rep.recog_repetition_text(self, show)
df = self.compos_dataframe
df = df.merge(df_nontext, how='left')
df.loc[df['alignment'].isna(), 'alignment'] = df_text['alignment']
df = df.merge(df_text, how='left')
df.rename({'alignment': 'alignment_in_group'}, axis=1, inplace=True)
if clean_attrs:
df = df.drop(list(df.filter(like='cluster')), axis=1)
df = df.fillna(-1)
for i in range(len(df)):
if df.iloc[i]['group_nontext'] != -1:
df.loc[i, 'group'] = 'nt-' + str(
int(df.iloc[i]['group_nontext']))
elif df.iloc[i]['group_text'] != -1:
df.loc[i,
'group'] = 't-' + str(int(df.iloc[i]['group_text']))
groups = df.groupby('group').groups
for i in groups:
if len(groups[i]) == 1:
df.loc[groups[i], 'group'] = -1
df.group = df.group.fillna(-1)
# df = rep.rm_invalid_groups(df)
self.compos_dataframe = df
def repetitive_group_recognition(self,
show=False,
clean_attrs=True): # call from notebook
df_nontext = rep.recog_repetition_nontext(
self, show) #call this func recog_repetition_nontext
df_text = rep.recog_repetition_text(self, show)
#call this func recog_repetition_text
df = self.compos_dataframe #original data frame
df = df.merge(df_nontext, how='left')
#print(df) #merged with nontext
#print(df_text['alignment'])
if 'alignment' not in df.columns:
#print(True)
NaN = np.nan
df["alignment"] = NaN
# print(df)
# print("df_text")
# print(df_text)
df.loc[df['alignment'].isna(), 'alignment'] = df_text['alignment']
df = df.merge(df_text, how='left')
df.rename({'alignment': 'alignment_in_group'}, axis=1, inplace=True)
if clean_attrs:
df = df.drop(list(df.filter(like='cluster')), axis=1)
df = df.fillna(-1)
for i in range(len(df)):
if df.iloc[i]['group_nontext'] != -1:
df.loc[i, 'group'] = 'nt-' + str(
int(df.iloc[i]['group_nontext']))
elif df.iloc[i]['group_text'] != -1:
df.loc[i,
'group'] = 't-' + str(int(df.iloc[i]['group_text']))
groups = df.groupby('group').groups
for i in groups:
if len(groups[i]) == 1:
df.loc[groups[i], 'group'] = -1
df.group = df.group.fillna(-1)
# df = rep.rm_invalid_groups(df)
self.compos_dataframe = df
def repetitive_group_recognition(self, show=False, clean_attrs=True):
df_nontext = rep.recog_repetition_nontext(self, show)
df_text = rep.recog_repetition_text(self, show)
df = self.compos_dataframe
df = df.merge(df_nontext, how='left')
df.loc[df['alignment'].isna(), 'alignment'] = df_text['alignment']
df = df.merge(df_text, how='left')
if clean_attrs:
df = df.drop(list(df.filter(like='cluster')), axis=1)
df = df.fillna(-1)
for i in range(len(df)):
if df.iloc[i]['group_nontext'] != -1:
df.loc[i, 'group'] = 'nt-' + str(
int(df.iloc[i]['group_nontext']))
elif df.iloc[i]['group_text'] != -1:
df.loc[i,
'group'] = 't-' + str(int(df.iloc[i]['group_text']))
# df[list(df.filter(like='group'))] = df[list(df.filter(like='group'))].astype(int)
self.compos_dataframe = df