【量化分析】Demo版：获取stock市场往日数据并预测未来两周走势

加载工具库

import matplotlib.pyplot as plt
import numpy as np
import matplotlib.pyplot as plt
from pathlib import Path
from gluonts.model import deepar
from gluonts.dataset import common
from gluonts.dataset.util import to_pandas
from gluonts.model.predictor import Predictor
import gluonts
import mxnet as mx
import baostock as bs
import pandas as pd

数据下载函数

def getData(code='sh.000001'):
    #### 登陆系统 ####
    lg = bs.login()
    # 显示登陆返回信息
    print('login respond error_code:'+lg.error_code)
    print('login respond  error_msg:'+lg.error_msg)

    #### 获取沪深A股历史K线数据 ####
    # 详细指标参数，参见“历史行情指标参数”章节；“分钟线”参数与“日线”参数不同。
    # 分钟线指标：date,time,code,open,high,low,close,volume,amount,adjustflag
    # 周月线指标：date,code,open,high,low,close,volume,amount,adjustflag,turn,pctChg
    rs = bs.query_history_k_data_plus("sh.000001",
        "date,code,open,high,low,close,preclose,volume,amount,adjustflag,turn,tradestatus,pctChg,isST",
        start_date='2000-01-01', end_date='2020-09-26',
        frequency="d", adjustflag="3")
    print('query_history_k_data_plus respond error_code:'+rs.error_code)
    print('query_history_k_data_plus respond  error_msg:'+rs.error_msg)

    #### 打印结果集 ####
    data_list = []
    while (rs.error_code == '0') & rs.next():
        # 获取一条记录，将记录合并在一起
        data_list.append(rs.get_row_data())
    result = pd.DataFrame(data_list, columns=rs.fields)

    #### 结果集输出到csv文件 ####   
#     result.to_csv("%s.csv"%(code), index=False)

    #### 登出系统 ####
    bs.logout()
    return result

主要代码

data = getData()
value = data['close'].values
value = value.astype(np.float32)
date = data['date'].values
date = date.astype(np.str)

data_dict={'timestamp':date,'value':value}
df = pd.DataFrame(data=data_dict)
df.to_csv("data.csv", index=False)
data = pd.read_csv("data.csv", header=0, index_col=0)
training_data = common.ListDataset([{"start": data.index[0],
    "target": data.value[:"2020-09-18"]}], freq="d")
testing_data = common.ListDataset([{"start": data.index[0],
    "target": data.value[:"2020-09-18"]}], freq="d")

trainer=gluonts.trainer._base.Trainer(batch_size=32, 
                                      clip_gradient=10.0, 
                                      ctx=mx.gpu(0), 
                                      epochs=10, 
                                      hybridize=True, 
                                      init="xavier", 
                                      learning_rate=0.001, 
                                      learning_rate_decay_factor=0.5, 
                                      minimum_learning_rate=5e-05, 
                                      num_batches_per_epoch=len(value)//32,
                                      patience=10, 
                                      weight_decay=1e-05)
estimator = deepar.DeepAREstimator(freq="d", prediction_length=14,trainer=trainer)
predictor = estimator.train(training_data=training_data)

for test_entry, forecast in zip(testing_data, predictor.predict(testing_data)):
    to_pandas(test_entry)[-15:].plot(linewidth=2)
    forecast.plot(color='g', prediction_intervals=[50.0, 90.0])
plt.grid(which='both')
plt.show()

结果展示