Stock Trading Market OpenAI Gym Environment with PARL

Overview

base on :

Reference

[1] https://github.com/wangshub/RL-Stock

[2] https://github.com/kh-kim/stock_market_reinforcement_learning

[3] https://github.com/PaddlePaddle/Paddle

[4] https://github.com/PaddlePaddle/PARL

Requirements

• Python3.7
• Numpy
• parl
• paddle
• OpenAI Gym

GET DATA:

$ python get_stock_data.py

run:

    $ python DDPG_STOCK.py

$ python DQN_STOCK.py

###################################################################### ######################################################################

7. 请选择你训练的最好的一次模型文件做评估

###################################################################### ###################################################################### def stock_trade(stock_file,ckpt_files): ckpt = ckpt_files # 请设置ckpt为你训练中效果最好的一次评估保存的模型文件名称 agent.restore(ckpt) # 创建环境 df_test = pd.read_csv(stock_file) df_test = df_test.sort_values('date') # The algorithms require a vectorized environment to run env_test = StockTradingEnv(df_test) day_profits = [] # The algorithms require a vectorized environment to run obs = env_test.reset() for i in range(len(df_test) - 1): batch_obs = np.expand_dims(obs, axis=0) action = agent.predict(batch_obs.astype('float32')) action = np.squeeze(action)

    # 给输出动作增加探索扰动，输出限制在 [-1.0, 1.0] 范围内
    action = np.random.normal(action, 1.0)
    action = np.clip(action, -1.0, 1.0)
    # 动作映射到对应的 实际动作取值范围 内, action_mapping是从parl.utils那里import进来的函数
    action = action_mapping(action, env.action_space.low[0],
                            env.action_space.high[0])
    next_obs, reward, done,info= env.step(action)
    Net_worth = env.render()
    day_profits.append(Net_worth)
return day_profits

def test_a_stock_trade(stock_file,ckpt_files): daily_profits = stock_trade(stock_file,ckpt_files) daily_profits= pd.DataFrame(daily_profits) df_test = pd.read_csv(stock_file) df_test = df_test.sort_values('date') fig, ax = plt.subplots() x = df_test['date'].drop(0) y1 = daily_profits[0] ax.plot(x,y1) ax.grid() #ax.legend(prop=font) #plt.show() plt.savefig(f'test.png')