feat: hist and pie plot

Author: wangshub

.gitignore

@@ -5,6 +5,7 @@ stockdata
 model
 log
 .vscode/
+*.pkl
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

README.md

@@ -99,14 +99,37 @@ reward = 1 if reward > 0 else reward = -100
 
 ### 验证结果
 
-#### 单只股票测试
+**单只股票**
 
-#### 单只股票多次测试
+- 初始本金 `10000`
+- 股票代码：`sh.600036`(招商银行)
+- 训练集： `stockdata/train/sh.600036.招商银行.csv`
+- 测试集： `stockdata/test/sh.600036.招商银行.csv`
+- 模拟操作 `20` 天，最终盈利约 `400`
 
-#### 多只股票测试
+<img src="img/sh.600036.png" alt="drawing" width="70%"/>
 
-## 结语
+**多只股票**
 
-欢迎指正！
+选取 `79` 只股票，进行训练，共计 
+
+- 盈利： `44.8%`
+- 不亏不赚： `49.0%`
+- 亏损：`6.3%`
+
+<img src="img/pie.png" alt="drawing" width="60%"/>
+
+<img src="img/hist.png" alt="drawing" width="60%"/>
+
+## 最后
+
+- 俺完全是股票没入门的新手，难免存在错误，欢迎指正！
+- 数据和方法皆来源于网络，无法保证有效性，**Just For Fun**！
 
 ## 参考资料
+
+- Y. Deng, F. Bao, Y. Kong, Z. Ren and Q. Dai, "Deep Direct Reinforcement Learning for Financial Signal Representation and Trading," in IEEE Transactions on Neural Networks and Learning Systems, vol. 28, no. 3, pp. 653-664, March 2017.
+- [Yuqin Dai, Chris Wang, Iris Wang, Yilun Xu, "Reinforcement Learning for FX trading"](http://stanford.edu/class/msande448/2019/Final_reports/gr2.pdf)
+- [Create custom gym environments from scratch — A stock market example](https://towardsdatascience.com/creating-a-custom-openai-gym-environment-for-stock-trading-be532be3910e)
+- [notadamking/Stock-Trading-Environment](https://github.com/notadamking/Stock-Trading-Environment)
+- Chien Yi Huang. Financial trading as a game: A deep reinforcement learning approach. arXiv preprint arXiv:1807.02787, 2018.

img/hist.png

@@ -1,4 +1,5 @@
 import os
+import pickle
 import pandas as pd
 from stable_baselines.common.policies import MlpPolicy
 from stable_baselines.common.vec_env import DummyVecEnv
@@ -40,14 +41,15 @@ def stock_trade(stock_file):
 
 
 def find_file(path, name):
+    # print(path, name)
     for root, dirs, files in os.walk(path):
         for fname in files:
             if name in fname:
                 return os.path.join(root, fname)
 
 
 def test_a_stock_trade(stock_code):
-    stock_file = find_file('./stockdata/train', stock_code)
+    stock_file = find_file('./stockdata/train', str(stock_code))
 
     daily_profits = stock_trade(stock_file)
     fig, ax = plt.subplots()
@@ -61,11 +63,24 @@ def test_a_stock_trade(stock_code):
 
 
 def multi_stock_trade():
-    pass
+    start_code = 600000
+    max_num = 3000
+
+    group_result = []
+
+    for code in range(start_code, start_code + max_num):
+        stock_file = find_file('./stockdata/train', str(code))
+        if stock_file:
+            profits = stock_trade(stock_file)
+            group_result.append(profits)
+
+    with open(f'code-{start_code}-{start_code + max_num}.pkl', 'wb') as f:
+        pickle.dump(group_result, f)
 
 
 if __name__ == '__main__':
-    test_a_stock_trade('sh.600036')
+    multi_stock_trade()
+    # test_a_stock_trade('sh.600036')
     # ret = find_file('./stockdata/train', '600036')
     # print(ret)