search

sadighian / crypto-rl

Deep Reinforcement Learning toolkit: record and replay cryptocurrency limit order book data & train a DDQN agent
850 stars 235 forks source link

it should be have a2c.py ppo.py? #17

Closed playmyswift closed 4 years ago

playmyswift commented 4 years ago

thanks for you repo, I read your paper, the repo dose't have ppo.py? because .gitignore file include,

agent/dqn_weights/*
agent/ppo_weights/*
agent/acer_weights/*
agent/a2c_weights/*

https://github.com/sadighian/crypto-rl/commit/5cfbdf1d175fe7554e337337866bec82f73eca9b so, I want to replay your paper test, but I don't know how can create ppo acer a2c.py

playmyswift commented 4 years ago

and it seems dqn looks like the worst?

sadighian commented 4 years ago

Hi @ericjjj ,

This repo is not the source code to the papers, but a platform for people to accelerate their RL research (i.e., data collection and integration is already provided).

The DQN agent provided in this repo serves as a "hello world" example for platform integration with an RL algorithm. If you are interested in extending the platform to use other RL algos (i.e., PPO, ACER, A2C, ...), then it might be helpful to look at the source code in dqn.py.