Transforms historical data and indicator values into normalized values to be used by AI
This repo contains historical price data for the EURUSD pair in 3 timeframes (D1, H4, and H1). Use this site to download the most recent historical data.
How to run:
# make sure to put the data csv files in the correct order (D1, H4, then H1)
python preprocessor.py EURUSD_D1.csv EURUSD_H4.csv EURUSD_H1.csv
This program outputs 3 files with names ending with "_data.csv" that contain the preprocessed normalized values.
Preprocessed values:
# DATETIME
# OPEN
# HIGH
# LOW
# CLOSE
# CLOSE DERIVATIVE
# VOLUME
# VOLUME DERIVATIVE
# EMA(500)
# EMA(200)
# EMA(100)
# EMA(50)
# EMA(25)
# EMA(8)
# EMA(500) DERIVATIVE
# EMA(200) DERIVATIVE
# EMA(100) DERIVATIVE
# EMA(50) DERIVATIVE
# EMA(25) DERIVATIVE
# EMA(8) DERIVATIVE
# STDDEV(20)
# MACD_SIGNAL(12, 26, 9)
These set of values aim to recreate this trading environment for the AI:
