正常clone,在virtualenv的环境下pip install -r requirements.txt
然后好像还是缺了几个库,总之缺啥补啥。
Step 1: 下载数据
python main.py --mode=download_data
然后数据会下载到PGPortfolio/database的路径下,大概是一个160M左右的.db文件。
Step2: Training and Tuning the hyper-parameters
按照GitHub上的readme,操作流程大致如下:
First, change the content in the nntrader/nntrader/net_config.json file.
make sure current directory is under nntrader and type
python main.py --mode=generate --repeat=1
this will make 1 subfolders under the train_package
in each subfloder, there is a copy of the net_config.json
–repeat=n, n could follow any positive integers. The random seed of each the subfolder ranges from 0 to n-1.
type
python main.py --mode=train --processes=1
this will start training one b