def wordTuples(graph, textEntry):
text = rootify(graph, textEntry)
pt = textacy.load_spacy('pt')
doc = textacy.Doc(text, lang=pt)
ts = textacy.TextStats(doc)
words = [{
w[0]: w[1]
} for w in textacy.keyterms.textrank(
doc, normalize='lower', n_keyterms=ts.n_unique_words)]
return words
def __init__(self):
self._content = ""
self._min_occurrence_for_topic = 2
self._common_verbs = 10
# create an empty corpus
en = textacy.load_spacy('en_core_web_sm', disable=('parser',))
self._corpus = textacy.Corpus(lang=en)
self._content = None
self._model = None
self._numdocs = 0
self._numtopics = 0
self._terms = None
self._doc_term_matrix = None
self._doc_topic_matrix = None
self._vectorizer = Vectorizer(tf_type='linear', apply_idf=True, idf_type='smooth',
norm='l2', min_df=3, max_df=0.95, max_n_terms=100000)
def setup_nlp_language_model(language, **nlp_args):
if (len(nlp_args.get('disable', [])) == 0):
nlp_args.pop('disable')
def remove_whitespace_entities(doc):
doc.ents = [e for e in doc.ents if not e.text.isspace()]
return doc
logger.info('Loading model: %s...', language)
Language.factories[
'remove_whitespace_entities'] = lambda nlp, **cfg: remove_whitespace_entities
model_name = LANGUAGE_MODEL_MAP[language]
#if not model_name.endswith('lg'):
# logger.warning('Selected model is not the largest availiable.')
nlp = textacy.load_spacy(model_name, **nlp_args)
nlp.tokenizer = keep_hyphen_tokenizer(nlp)
pipeline = lambda: [x[0] for x in nlp.pipeline]
logger.info('Using pipeline: ' + ' '.join(pipeline()))
return nlp
# coding: utf-8
# In[65]:
from cassandra.cluster import Cluster
import spacy
import pandas as pd
import operator
from datetime import datetime
import textacy
nlp = spacy.load('en_core_web_lg')
cluster = Cluster(['127.0.0.1'])
session = cluster.connect('sharelock')
en = textacy.load_spacy('en_core_web_lg')
# Clean text rows, lemmatize, remove stop words.
#
# In[66]:
from emoji import UNICODE_EMOJI
def is_emoji(s):
count = 0
for emoji in UNICODE_EMOJI:
count += s.count(emoji)
if count > 1:
return False
return bool(count)
# pip install textblob
# python -m textblob.download_corpora
# pip install spacy textacy
# python -m spacy download en
import json
from collections import defaultdict, Counter
import re
import textacy
from textblob import TextBlob
en = textacy.load_spacy("en_core_web_sm")
# my_stop_words = {"the", "to"}
my_stop_words = en.Defaults.stop_words - {"not"}
data = []
with open("only_cellphones_brands.json") as json_data:
data = json.load(json_data)
# re-implemented from https://github.com/chartbeat-labs/textacy/blob/26abafae7d745614b68d2f90310741b5a8de24d7/textacy/extract.py#L327
# to ignore stop words in matching
def pos_regex_matches(doc, pattern):
# standardize and transform the regular expression pattern...
pattern = re.sub(r"\s", "", pattern)
pattern = re.sub(r"<([A-Z]+)\|([A-Z]+)>", r"( (\1|\2))", pattern)
pattern = re.sub(r"<([A-Z]+)>", r"( \1)", pattern)
filtered_tokens = [
tok for tok in doc if tok.text not in my_stop_words and tok.is_alpha
import pickle as pkl
import re
import numpy as np
import pandas as pd
from gensim.models.phrases import Phrases, Phraser
import textacy
from textacy import preprocess_text, Doc, Corpus
from textacy.vsm import Vectorizer, GroupVectorizer
from textacy.tm import TopicModel
en = textacy.load_spacy("en_core_web_sm", disable='parser')
data = pd.read_csv('data/qaData.csv', parse_dates=['Date'])
ec_data = data.loc[data['EventType'] == "Earnings call", [
'Date', 'Company', 'Participants', 'AnalystName', 'AnalystCompany',
'EventName', 'EarningTag2', "Question"
]].copy()
ec_data['Quarter'] = ec_data['EventName'].str.split("Q").str[0]
ec_data = ec_data.groupby(
['Date', "Company", "Participants", "EventName",
"Quarter"]).apply(lambda x: x.reset_index()).reset_index(drop=True)
ec_data.columns = [
"QuestionOrder", "Date", "Company", "Participants", "AnalystName",
"AnalystCompany", "EventName", "Tag", "Question", "Quarter"
]
ec_data = ec_data[[
"Date", "Quarter", "Company", "Participants", "AnalystCompany",
"AnalystName", "QuestionOrder", "Tag", "Question"
]]
import spacy
import textacy
from spacy import displacy, tokens
from textacy import extract
# This is common global object which is used for computing similarity
# Some private members so as to make it work
nlp = spacy.load('en_core_web_sm')
lang_en = textacy.load_spacy('en_core_web_sm')
possible_docs = {'textacy', 'spacy'}
subjects = ["nsubj", "nsubjpass", "csubj", "csubjpass", "agent", "expl"]
objects = ["dobj", "dative", "attr", "oprd"]
conjunctions = ["or", "and"]
def render_pos_html(list_of_docs):
return displacy.render(map(lambda x: get_spacy_doc(x), list_of_docs),
style='dep',
page=True)
def get_new_doc(phrase, doc_type='textacy'):
assert isinstance(phrase, basestring)
assert isinstance(doc_type, str)
assert doc_type in possible_docs, "Only {} doc types are supported".format(
possible_docs)
if doc_type == 'textacy':
return textacy.Doc(phrase, lang=lang_en)
elif doc_type == 'spacy':
return nlp(phrase)