import pandas as pd
import numpy as np
import io
import requests
from io import StringIO

import sklearn
import sklearn.ensemble
import sklearn.metrics
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import *
from sklearn.utils import shuffle


url = "https://storage.googleapis.com/tensorflow-workshop-examples/stack-overflow-data.csv" # .csv file location
s = requests.get(url).content
df = pd.read_csv(io.StringIO(s.decode('utf-8')))

df = df[pd.notnull(df['tags'])]
df = df.sample(frac=0.5, random_state=99).reset_index(drop=True)
df = shuffle(df, random_state=22)
df = df.reset_index(drop=True)
df['class_label'] = df['tags'].factorize()[0]


df_train, df_test = train_test_split(df, test_size=0.2, random_state=40)

X_train = df_train["post"].tolist()
X_test = df_test["post"].tolist()
y_train = df_train["class_label"].tolist()
y_test = df_test["class_label"].tolist()

vectorizer = CountVectorizer(analyzer='word',token_pattern=r'\w{1,}', ngram_range=(1, 3), stop_words = 'english', binary=True)
train_vectors = vectorizer.fit_transform(X_train)
test_vectors = vectorizer.transform(X_test)

logreg = LogisticRegression(n_jobs=1, C=1e5)
logreg.fit(train_vectors, y_train)