| Time interval | ROI |
|---|---|
| 1d (Human) | 2.74% |
| 1d (Model) | 304.96% |
| 4h (Human) | 36.86% |
| 4h (Model) | 405.03% |
| 1h (Human) | 37.55% |
| 1h (Model) | 290.01% |
| Time interval | ROI |
|---|---|
| 1d (Human) | 3.11% |
| 1d (Model) | 12.19% |
| 4h (Human) | 18.30% |
| 4h (Model) | 31.64% |
| 1h (Human) | 19.79% |
| 1h (Model) | 27.58% |
- Test OS: Ubuntu 16.04 LTS
- Python version: 3.8
- Create folders.
mkdir images
mkdir checkpoints- Please run
pip install –r requirements.txtto install the needed libraries.
- Clone the repo.
- Follow the instruction to download required data.
# ETHUSDT
python download-kline.py -s ETHUSDT -startDate 2017-08-01 -endDate 2021-12-01
# BTCUSDT
python download-kline.py -s BTCUSDT -startDate 2017-08-01 -endDate 2021-12-01- It will download the required data as below. Unzip the zip files under the
1h,4hand1ddirectories.
binance_prediction_pytorch
`-- binance-public-data
`-- data
`-- data
`-- spot
|-- daily
`-- monthly
`-- klines
|-- ETHUSDT
`-- BTCUSDT- Then soft link the data directory to the repo root as below.
binance_prediction_pytorch
|-- binance-public-data
`-- data
`-- spot
|-- daily
`-- monthly
`-- klines
|-- ETHUSDT
`-- BTCUSDT- Run training and evaluation on ETHUSDT. It will store the checkpoints under
checkpointswith ticker name and time interval if don't specify the checkpoint path with--ckpt.
# 1d
./run.sh ETHUSDT 1d
# 4h
./run.sh ETHUSDT 4h
# 1h
./run.sh ETHUSDT 1h- Run training and evaluation on BTCUSDT
# 1d
./run.sh BTCUSDT 1d
# 4h
./run.sh BTCUSDT 4h
# 1h
./run.sh BTCUSDT 1h- Specify the checkpoint path with
evalmode to only do the inference.
./run.sh ETHUSDT 1h --ckpt ${YOUR_CHECKPOINT_PATH} --eval- Directly try with jupyter notebook file Final.ipynb.