From af6525f106fd2a79c3018b050183fc69f8716281 Mon Sep 17 00:00:00 2001 From: Mingxin WANG <37770806+kzil88@users.noreply.github.com> Date: Mon, 23 Apr 2018 09:57:10 +0800 Subject: [PATCH 1/3] Add files via upload --- BP.py | 42 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) create mode 100644 BP.py diff --git a/BP.py b/BP.py new file mode 100644 index 0000000..480636a --- /dev/null +++ b/BP.py @@ -0,0 +1,42 @@ +from keras.models import Sequential +from keras.layers import Dense, Dropout, Activation +from keras.optimizers import SGD +import DC +from keras.layers import Flatten +import numpy as np +import keras + + +if __name__ == '__main__': + dc = DC.DC('600519') + score_list = [] + resu_list = [] + train = dc.data_train + target = dc.data_target + + + model = Sequential() + model.add(Dense(64, activation='linear', input_dim=13)) + model.add(Dropout(0.5)) + model.add(Dense(64, activation='sigmoid')) + model.add(Dropout(0.5)) + model.add(Dense(1, activation='relu')) + sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True) + model.compile(loss='logcosh', optimizer=sgd, metrics=['accuracy']) + + + for i in range(5): + model.fit(train, target, epochs=20000) + score = model.evaluate(train, target, batch_size=128) + print('SCORE:' + str(score[0])) + test_case = np.array([dc.test_case]) + ans2 = model.predict(test_case) + resu_list.append(ans2[0][0]) + score_list.append(score) + print('RESU '+str(i+1)+' : '+str(ans2[0][0])) + dc.refreshDATA(ans2[0][0]) + train = dc.data_train + target = dc.data_target + + print(score_list) + print(resu_list) From e58cc68a3e987af191150be485ae4fd66c144acc Mon Sep 17 00:00:00 2001 From: Mingxin WANG <37770806+kzil88@users.noreply.github.com> Date: Mon, 23 Apr 2018 10:21:08 +0800 Subject: [PATCH 2/3] Update README.md --- README.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/README.md b/README.md index 16e598b..2ec58d5 100644 --- a/README.md +++ b/README.md @@ -19,3 +19,7 @@ Init_Profit.py ------ update the profit and divide shares of stock on Market. Init_RZRQ.py ------ update the RZRQ information of stock on Market. Init_RZRQ_Detail.py ------ update the RZRQ detail of stock on Market. Init_Single.py ------ update specific stock daily price information on Market. + +---------------------------------------------------------------------------------- + +BP.py ------ update Q1.BP.py to branch Q1. From 5edd3c6ae50e46c5199a1a1f510a6ede7def3fad Mon Sep 17 00:00:00 2001 From: Mingxin WANG <37770806+kzil88@users.noreply.github.com> Date: Tue, 24 Apr 2018 13:58:22 +0800 Subject: [PATCH 3/3] Add files via upload --- ChooseStock.py | 161 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 161 insertions(+) create mode 100644 ChooseStock.py diff --git a/ChooseStock.py b/ChooseStock.py new file mode 100644 index 0000000..405124d --- /dev/null +++ b/ChooseStock.py @@ -0,0 +1,161 @@ +# -*- coding:utf8 -*- +import tushare as ts +import re +import numpy as np +from pandas import DataFrame +import matplotlib.pyplot as plt +import pandas as pd +import datetime +import talib as ta +import random +import pymysql.cursors + + + +class CS(object): + code = '' + avg = [] + good_factor = 0.02 + bad_factor = 0.05 + cnt_bad_sell = 0 + cnt_good_buy = 0 + cnt_good_sell = 0 + cnt_risk = 0 + af = [] + fq = [] + process = [] + dd_list = [] + dd_list_show = [] + macd_list = [] + kdj_list = [] + bool_up = 0.00 + bool_mid = 0.00 + bool_dn = 0.00 + data_train = [] + data_target = [] + close = [] + test_case = [] + + def __init__(self, in_code): + self.collectDATA(in_code) + + def collectDATA(self,in_code): + self.code = in_code + + self.data_src = temp_day = ts.get_k_data(code=in_code) + if temp_day.shape[0] < 150 : + raise Exception + list_open = np.array(temp_day.iloc[temp_day.shape[0]-20:temp_day.shape[0], [1]].astype('float32'), dtype=np.float).ravel() + list_close = np.array(temp_day.iloc[temp_day.shape[0]-20:temp_day.shape[0], [2]].astype('float32'), dtype=np.float).ravel() + list_high = np.array(temp_day.iloc[temp_day.shape[0]-20:temp_day.shape[0], [3]].astype('float32'), dtype=np.float).ravel() + list_low = np.array(temp_day.iloc[temp_day.shape[0]-20:temp_day.shape[0], [4]].astype('float32'), dtype=np.float).ravel() + self.list_vol = np.array(temp_day.iloc[temp_day.shape[0]-20:temp_day.shape[0], [5]].astype('float32'), dtype=np.float).ravel() + period = 20 + self.close = list_close + self.avg = ta.MA(list_close,period) + self.avg = [x for x in self.avg if str(x) != 'nan'] + self.good_buy = [x* (1.00 - self.good_factor) for x in self.avg] + self.good_sell = [x * (1.00 + self.good_factor) for x in self.avg] + self.bad_sell = [x * (1.00 - self.bad_factor) for x in self.avg] + self.cnt_risk = [0]*len(self.avg) + self.cnt_good_sell = [0]*len(self.avg) + self.cnt_good_buy = [0]*len(self.avg) + self.cnt_bad_sell = [0]*len(self.avg) + for a in range(len(self.avg)): + self.cnt_bad_sell[a] = len([x for x in list_low[:a+period-1] if x <= self.bad_sell[a]]) + self.cnt_good_sell[a] = len([x for x in list_high[:a + period - 1] if x >= self.good_sell[a]]) + self.cnt_bad_sell[a] = len([x for x in list_low[:a + period - 1] if self.bad_sell[a] < x <= self.good_buy[a]]) + self.cnt_risk[a] = len([x for x in list_low[:a+period-1] if x <= list_close[a+period-1]]) + + #ARFQ + for b in range(len(self.avg)): + af,fq,process = get_arfq(list_high[b:b+period],list_low[b:b+period],self.good_sell[b],self.bad_sell[b],self.good_buy[b]) + self.af.append(af/self.avg[-1]) + self.fq.append(fq) + self.process.append(process) + + + +def get_arfq(list_high,list_low,good_sell,bad_sell,good_buy): + # 振幅af = (high-low)/len 频率 freq = 从good_selld到good_buy（或反之）的所需步长之和除以len + af = ((list_high - list_low).sum()) / (len(list_high)) + start_flag = 0 + list_index = 0 + #list_high = [float(x) for x in list_high] + #list_low = [float(x) for x in list_low] + for k in range(len(list_high)): + if list_high[len(list_high) - k - 1] >= good_sell: + start_flag = 1 + list_index = len(list_high) - k - 1 + elif bad_sell < list_low[len(list_high) - k - 1] <= good_buy: + start_flag = 2 + list_index = len(list_high) - k - 1 + freq_list = [] + freq = 0 + freq_step = [] + process = 0 + for l in range(list_index, len(list_high)): + if start_flag > 0: + if (divmod(start_flag, 2)[1] == 1) and (bad_sell < list_low[l] <= good_buy): + freq_list.append(l) + start_flag = start_flag + 1 + elif (divmod(start_flag, 2)[1] == 0) and (list_high[l] >= good_sell): + freq_list.append(l) + start_flag = start_flag + 1 + else: + freq = 0 + if len(freq_list) == 1: + freq = 0 - (len(list_high) - freq_list[0] - 1) + elif len(freq_list) > 1: + for m in range(1, len(freq_list)): + freq_step.append(freq_list[len(freq_list) - m] - freq_list[len(freq_list) - 1 - m]) + freq = np.array(freq_step).sum() / (len(freq_step)) + + if freq > 0: + process = (len(list_high) - freq_list[-1] - 1) / (freq) + else: + process = 0 + return af,freq,process + +if __name__ == '__main__': + # 取股票代码清单 + src = ts.get_stock_basics() + code_list = src.index + code_set = set() + for i in range(len(code_list)): + resu = re.search(r'\d+', code_list[i]).string + code_set.add(resu) + print('Done:Code_Set') + + # 建立数据库连接 + db = pymysql.connect(host='127.0.0.1', user='root', passwd='admin', db='stock', charset='utf8') + cursor = db.cursor() + sql_done_set = "SELECT distinct stock_code FROM q1_cs" + try: + cursor.execute(sql_done_set) + done_set = cursor.fetchall() + for row in done_set: + code_set.remove(row[0]) + + cnt = 1 + total = len(code_set) + for i in code_set: + print(i) + try: + cs = CS(i) + sql = "INSERT INTO q1_cs(stock_code,af,freq,process) VALUES ('%s', '%f', '%f','%f')" % (i, cs.af[-1], cs.fq[-1], cs.process[-1]) + cursor.execute(sql) + db.commit() + cnt += 1 + print('Seq: '+str(cnt)+' of ' + str(total) +'Code: ' + str(i) + ' AF: ' + str(cs.af[-1]) + ' Freq: ' + str(cs.fq[-1]) + ' Process: ' + str(cs.process[-1])) + except Exception as excp: + # db.rollback() + print(str('Errr') + str(i)) + print('All Finished!') + + except Exception as excp: + #db.rollback() + print(str('Errr')+ str(i)) + + db.close() +