def extract_users (self, calendar_df_iterator):
"""
given an iterator over calendar dataframes,
this constructs and returns a dataframe
containing all users
"""
print_header ("EXTRACTING USERS")
#==========[ ITERATE OVER ALL DFS ]==========
for cdf in calendar_df_iterator ():
print_status ("Extract users", "next df")
#=====[ Step 1: sort by user ]=====
print_inner_status ("extract_users", "sorting by user id")
cdf = cdf.sort ('user')
#=====[ Step 2: init user representations ]=====
print_inner_status ("extract_users", "initializing user representations")
unique_uids = [uid for uid in cdf['user'].unique ()]
for uid in unique_uids:
if not uid in self.user_representations:
self.user_representations[uid] = self.init_user_representation(uid)
#=====[ Step 3: update the user representations ]=====
print_inner_status ("extract_users", "updating user representations")
cdf.apply (self.update_user_representation, axis = 1)
#=====[ Step 4: convert to df, delete irrelevant stuff ]=====
print_inner_status ("extract_users", "converting to dataframe")
self.users_df = pd.DataFrame(self.user_representations.values())
del self.user_representations
return self.users_df
def get_corpus_dictionary (self):
"""
PRIVATE: get_corpus_dictionary
------------------------------
Assembles a gensim corpus and dictionary from activities_df,
where each text is name || words.
"""
#=====[ Step 1: iterate through all activity dataframes ]=====
print_status ("get_corpus", "assembling texts")
texts = []
for df in self.storage_delegate.iter_activity_dfs ():
print_inner_status ("assembling texts", "next df")
texts += list(df.apply(self.extract_text, axis=1))
#=====[ Step 3: get dictionary ]=====
print_status ("get_corpus", "assembling dictionary")
dictionary = gensim.corpora.Dictionary(texts)
#=====[ Step 4: get corpus ]=====
print_status ("get_corpus", "assembling corpus")
corpus = [dictionary.doc2bow (text) for text in texts]
return corpus, dictionary
