Use Keras, Tensorflow and LSTM to predict stock price
Make sure you have Python, Tensorflow and Keras installed
- Tensorflow installation guide https://www.tensorflow.org/install/
- Keras documentation https://keras.io/
Once the environment is ready, just simply open a command prompt and type in python multiinputs.py
I treat it as a supervised training problem, the csv file contains multiple columns e.g. ['Open','High','Low','Close','Adj close', 'Volume']
,
the process will preprocess the data by appending the next day's columns to the right as the answer
and use LSTM
to work it out.
The data will be split to 90%/10% for training and evaluation, here is the example result