Visual explanation of the importance of examining your data before trying to predict / run an algorithm.
@JayMahabal / @H2Oai
Forked from 1wheel's block: you-draw-it
license: mit |
/* Setting things up. */ | |
var path = require('path'), | |
express = require('express'), | |
app = express(), | |
Twit = require('twit'), | |
config = { | |
/* Be sure to update the .env file with your API keys. */ | |
twitter: { | |
consumer_key: process.env.CONSUMER_KEY, | |
consumer_secret: process.env.CONSUMER_SECRET, |
from sklearn.externals import joblib | |
import tweepy | |
import json | |
def predict(tweet_text): | |
# Load our model and vectorizer | |
clf = joblib.load('model.pkl') | |
vectorizer = joblib.load('vectorizer.pkl') | |
# Make the retweet count prediction after vectorizing the tweet |
Visual explanation of the importance of examining your data before trying to predict / run an algorithm.
@JayMahabal / @H2Oai
Forked from 1wheel's block: you-draw-it
<!Doctype HTML> | |
<head> | |
<!-- <script src="d3.js"></script> --> | |
<script src="https://cdnjs.cloudflare.com/ajax/libs/d3/4.4.1/d3.js"></script> | |
<style> | |
.triangle { |
license: mit | |
height: 1000 | |
scrolling: no | |
border: |
license: mit | |
height: 600 |