ciurana2016 / predict_stock_py Goto Github PK

View Code? Open in Web Editor NEW

187.0 20.0 84.0 3 KB

This is a submission for the "Predicting Stock Prices challenge" by @Sirajology on Youtube. [runner-up]

Python 100.00%

predict_stock_py's Introduction

predict_stock_py

This is a submission for the "Predicting Stock Prices challenge" by @Sirajology on Youtube.

Overview

The python script "predict_stock.py" does the following:

Asks the user for a stock quote from NASDAQ (e.j: AAPL, FB, GOOGL)
Uses Tweepy to retrieve tweets about that stock.
Uses TextBlob to determine if the majority of the tweets are positive using sentiment analisys.
If the last is True, downloads the last year of prices for that stock, and trains a neural net with that data to predict the price for tomorrow.

The folder "demo" contains a test training 'Table 3' to the same network that is used to predict the price.

Dependencies

numpy (http://www.numpy.org/)
tweepy (http://www.tweepy.org/)
textblob (https://textblob.readthedocs.io/en/dev/)
requests(http://docs.python-requests.org/en/master/)
keras(https://keras.io/) Runs with TensorFlow or Theano, so you will need one of them.

Usage

Install all the necesary dependencies. Then just run:

python predict_stock.py

It will ask you for a NASDAQ quote, e.j: AAPL, then if the sentiment is positive and the stock you entered exists it will start training the network and give you a result.

Credits

Credits to Siraj and to this blog post.

Disclaimer

Do not use this code to invest in the stock market, if you are interested in stocks start by reading "The Intelligent Investor" by Benjamin Graham.

predict_stock_py's People

Contributors

Stargazers

Watchers

Forkers

satroan abishekpuri dejanrizvan dburner spkprav thedarsideofit abhishekkodi gandaqian bigandrewgold sumitneup jgabriellima chirbi scripnichenko zhangzexi hdasappinc lulzzz michaelluk sushantraikar rohithzr iabhishekofficial kinergy akbarboghani pliashkou psyclopz exceltior gulsunde pierrickvoulet yoka1985 kenken64 miatoro jtn-ms isasgeir sbusso tianhm blsh jimkliew gustianom owsas rakshakhegde ajinkyat mutkan brunocrt zikmout engahmed1190 nan1488 coventryresearch saadmahboob renews scryptolabs harshit1504 guoxiaoyong akram1026 arondalton nimmiw eltoro13 logbie matce cldmello alamops ugwakocrosky mukherjjee lslucas donadonny datapython mrmurga vicperotti nick-choudhary megaforce2 garvitburad anuragisi constantsojourner a20180502 evilluo mrmushfiq kailigu ppvastar scoobydopapa sgaikar1 ficuschain yazington krusty45 rohanlondhe123 hell-to-heaven

predict_stock_py's Issues

Details required

@ciurana2016 I am new to python and also for machine learning. Please bear with me.
I have few questions, if you can answer.

Did you test this with real world scenario? what was the success rate?
Why are you predicting the next sequence value with only "Open" value, is there any possibility that I can use multiple input and then predict the next sequence?

How do I setup and run

I tried running this, but I am getting error.
Can you guide me in the right way to install this?
Or a requirement.txt file will help too.

So, I tried it with several quotes and it runs fine, but for some quotes, it doesn't work properly. For instance, in the case of $EARS stock, it gives values that the stock was never priced it ever. Also, when I plugin quotes like $NAKD, I get the following error:

Using TensorFlow backend.
Enter a stock quote from NASDAQ (e.j: AAPL, FB, GOOGL): NAKD
Epoch 1/200
Traceback (most recent call last):
  File "predict_stock.py", line 105, in <module>
    print stock_prediction()
  File "predict_stock.py", line 81, in stock_prediction
    model.fit(trainX, trainY, nb_epoch=200, batch_size=2, verbose=2)
  File "/usr/local/lib/python2.7/dist-packages/keras/models.py", line 672, in fit
    initial_epoch=initial_epoch)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 1196, in fit
    initial_epoch=initial_epoch)
  File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 911, in _fit_loop
    callbacks.on_epoch_end(epoch, epoch_logs)
  File "/usr/local/lib/python2.7/dist-packages/keras/callbacks.py", line 76, in on_epoch_end
    callback.on_epoch_end(epoch, logs)
  File "/usr/local/lib/python2.7/dist-packages/keras/callbacks.py", line 265, in on_epoch_end
    self.progbar.update(self.seen, self.log_values, force=True)
AttributeError: 'ProgbarLogger' object has no attribute 'log_values

Please note: I don't put $ before the quotes.

Chance to Currency

Hello,
is it possible to adjust your script to get prediction for currencys insdead of stocks?
i tried to change the URL from:

https://finance.google.com/finance?q=NASDAQ%3A

to
https://finance.google.com/finance?q=CURRENCY%3A

As input i tried USD and EUR but no luck getting plenty of errors .

Also getting the error: "...bad sentiment"

ModuleNotFoundError: No module named 'tensorflow'

I tried running the script and i get this error. please help

Using TensorFlow backend.
Traceback (most recent call last):
File "E:\SM8\Projects\Stock Predictor\stock_predict.py", line 7, in
from keras.models import Sequential
File "C:\Users\shreyash\AppData\Local\Programs\Python\Python36\lib\site-packages\keras_init_.py", line 2, in
from . import backend
File "C:\Users\shreyash\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\backend_init_.py", line 67, in
from .tensorflow_backend import *
File "C:\Users\shreyash\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\backend\tensorflow_backend.py", line 1, in
import tensorflow as tf
ModuleNotFoundError: No module named 'tensorflow'