flair_predictor.py

import joblib
import nltk
from nltk.corpus import stopwords
import praw
from RedditAPI import accinfo
import pandas as pd
import sklearn
nltk.download('punkt')
info = accinfo()
stops = set(stopwords.words("english"))
model = joblib.load("./Models/finalized_model.sav")
reddit = praw.Reddit(client_id=info[0], client_secret=info[1], user_agent=info[2], username=info[3], password=info[4])

def title_preprocessing(row):
	title = row['title']
	tokens = nltk.word_tokenize(title)
	token_words = [w for w in tokens if w.isalnum()]
	key_words = [word for word in token_words if not word in stops]
	joined = (" ".join(key_words))
	return(joined)

def body_preprocessing(row):
	body = row['body']
	tokens = nltk.word_tokenize(body)
	token_words = [w for w in tokens if w.isalnum()]
	key_words = [W for W in token_words if not W in stops]
	joined = (" ".join(key_words))
	return(joined)
	

def comment_preprocessing(row):
	comments = row['comments']
	tokens = nltk.word_tokenize(comments)
	token_words = [w for w in tokens if w.isalnum()]
	key_words = [W for W in token_words if not W in stops]
	joined = (" ".join(key_words))
	return(joined)

def FlairifyMe(url):
	c = []
	s = ''
	post_dict = {"id": [], "title":[], "body":[], "comments": []}

	submission = reddit.submission(url = url)
	post_dict['id']=submission.id
	post_dict['title']=submission.title
	post_dict['body']=submission.selftext
	submission.comments.replace_more(limit=0)
	for comment in submission.comments.list():
		c.append(comment.body)
	post_dict['comments'].append(c)

	data = pd.DataFrame(post_dict)
	data.fillna("")

	data['title'] = data['title'].str.lower()
	data['body'] = data['body'].str.lower()
	for a in data['comments'][0]:
		s+=str(a)+" "
	data['comments']=s

	data['title_words'] = data.apply(title_preprocessing,axis=1)
	data['body_words'] = data.apply(body_preprocessing,axis=1)
	data['comment_words'] = data.apply(comment_preprocessing,axis=1)

	combine = data['title_words']+data['body_words']+data['comment_words']
	data = data.assign(combined=combine)
	return(model.predict(data['combined'])[0])
	
def FlairDetect(url):
	submission = reddit.submission(url = url)
	actual_flair = submission.link_flair_css_class
	return(actual_flair)