From 335290556fa90479b39e2850a9fdee4871e82b83 Mon Sep 17 00:00:00 2001 From: zhoushimin Date: Wed, 2 Nov 2022 16:48:51 +0800 Subject: [PATCH] YaoLowCodeDemo --- demo-widget | 1 + fund_tools.py | 730 -------------------------------------------------- start.sh | 7 + 3 files changed, 8 insertions(+), 730 deletions(-) create mode 160000 demo-widget delete mode 100644 fund_tools.py create mode 100755 start.sh diff --git a/demo-widget b/demo-widget new file mode 160000 index 0000000..2489060 --- /dev/null +++ b/demo-widget @@ -0,0 +1 @@ +Subproject commit 2489060e09630dff334e88b8e51e1efa1bfa1989 diff --git a/fund_tools.py b/fund_tools.py deleted file mode 100644 index 28123f9..0000000 --- a/fund_tools.py +++ /dev/null @@ -1,730 +0,0 @@ -from asyncio.windows_events import NULL -from cProfile import label -import akshare as ak -import pandas as pd -import numpy as np -from matplotlib import pyplot as plt -from matplotlib.pyplot import MultipleLocator -import matplotlib.dates as mdate -import requests -from bs4 import BeautifulSoup -import datetime -import time -import math -import os -import re -import random -from decimal import Decimal - -dir_cumulative_net_value_trend = 'cumulative_net_value_trend' -dir_cumulative_return_trend = 'cumulative_return_trend' - -# 基金行业常用常量定义 -# 无风险年化收益率% -risk_free_annual_return_ratio = 0.0275 -# 每年交易日天数 -trading_days_per_year = 250 - -# user_agent列表 -user_agent_list = [ - 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/21.0.1180.71 Safari/537.1 LBBROWSER', - 'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; QQDownload 732; .NET4.0C; .NET4.0E)', - 'Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.84 Safari/535.11 SE 2.X MetaSr 1.0', - 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Maxthon/4.4.3.4000 Chrome/30.0.1599.101 Safari/537.36', - 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.122 UBrowser/4.0.3214.0 Safari/537.36' -] - -# referer列表 -referer_list = [ - 'http://fund.eastmoney.com/110022.html', - 'http://fund.eastmoney.com/110023.html', - 'http://fund.eastmoney.com/110024.html', - 'http://fund.eastmoney.com/110025.html' -] - -# 获取一个随机user_agent和Referer -header = {'User-Agent': random.choice(user_agent_list), - 'Referer': random.choice(referer_list) -} - -# 获得字符串中的整数 -def get_decimal(x): - return ''.join(re.findall(r"\d+\.?\d*", ''.join(x))) - -# 获得指定后缀字符串中的整数 -def get_decimal_suffix(x, suffix): - return get_decimal(re.findall(r"\d+\.?\d*"+suffix, ''.join(x))) - -# 日志调试 -def log(x): - file_handle = open('log.txt', mode='a', encoding='utf-8') - file_handle.write(x) - file_handle.close() - -# 获得基金概要数据 -def get_all_fund_outline(): - if not os.path.exists('fund_em_fund_name_df.csv'): - df1 = ak.fund_name_em() - df1 = df1[['基金代码', '拼音缩写', '基金类型']] - df2 = ak.fund_open_fund_daily_em() - df = pd.merge(df1, df2, on='基金代码') - df.set_index(df['基金代码']) - df.to_csv('fund_em_fund_name_df.csv', encoding='utf_8_sig',index=None) - - print(df) - else: - df = pd.read_csv('fund_em_fund_name_df.csv', dtype=object) - - return df - -#获得公墓基金每日净值 -def get_all_open_fund_daily(): - if not os.path.exists('open_fund_daily.csv'): - fund_em_fund_name_df = ak.fund_open_fund_daily_em() - fund_em_fund_name_df.set_index(fund_em_fund_name_df['基金代码']) - fund_em_fund_name_df.to_csv('fund_em_open_fund_daily.csv', encoding='utf_8_sig',index=None) - else: - fund_em_fund_name_df = pd.read_csv('fund_em_open_fund_daily.csv', dtype=object) - - return fund_em_fund_name_df - -# 根据关键字过滤基金列表 -def query_fund_by_fundname_keyword(df, fundname_keyword): - # if not os.path.exists(fundname_keyword + '.csv'): - df_query = df[df['基金简称'].str.contains(fundname_keyword)] - df_query.set_index(df_query['基金代码']) - df_query.to_csv(fundname_keyword + '.csv', encoding='utf_8_sig',index=None); - # else: - # df_query = pd.read_csv(fundname_keyword + '.csv', dtype=object) - - return df_query - -# 获得分红数据 -def get_fund_dividend_by_id(dir, fund_id): - filename = dir + '/%s' % (fund_id) + '-dividend.csv' - try: - if not os.path.exists(filename): - df = ak.fund_open_fund_info_em(fund=fund_id, indicator="分红送配详情") - print(df) - for i in np.arange(0, df.shape[0]): - df.iloc[i]['每份分红'] = get_decimal(df.iloc[i]['每份分红']) - - df.to_csv(filename, encoding='utf_8_sig', index=None); - else: - df = pd.read_csv(filename, dtype=object) - - return df - except: - return NULL - - -# 获得最近一次分红 -def query_last_dividend_before_date(df, cur_date): - try: - dividend = '0.0' - # 查询最近一次的分红数据，满足条件反馈分红 - for i in np.arange(0, df.shape[0]): - if str(cur_date) > df.iloc[df.shape[0] - 1 - i]['分红发放日']: - # print('Index:', df.shape[0] - 1 - i) - dividend = df.iloc[df.shape[0] - 1 - i]['每份分红'] - - # print("日期:%s"%cur_date, "分红%s"%dividend) - - return dividend - except: - return '0.0' - -# 获得指定基金历史数据 -def get_fund_his_by_id(fund_id, dir): - filename = dir + '/%s'%(fund_id) + '.csv' - # print(filename) - - try: - if not os.path.exists(filename): - df_total_net_value_trend = ak.fund_open_fund_info_em(fund=fund_id, indicator='累计净值走势') - df_unit_net_value_trend = ak.fund_open_fund_info_em(fund=fund_id, indicator="单位净值走势") - fund_em_info_df = pd.merge(df_total_net_value_trend, df_unit_net_value_trend, on='净值日期') - fund_em_info_df.set_index(fund_em_info_df['净值日期']) - - if not isinstance(fund_em_info_df['日增长率'], float): - fund_em_info_df['日增长率'] = fund_em_info_df['日增长率'].astype(float) - - df_dividend = get_fund_dividend_by_id(dir, fund_id) - - # 复权列表 - red_net_value_list = [] - last_red_net_value = 0.0 - - # print(fund_em_info_df) - # print(f"fund_em_info_df.shape:{fund_em_info_df.shape}") - - # 复权净值计算 - for i in np.arange(0, fund_em_info_df.shape[0]): - dividend = float(query_last_dividend_before_date(df_dividend, fund_em_info_df.iloc[i]['净值日期'])) - # print(dividend) - - if dividend > 0.0: - try: - last_red_net_value = float(last_red_net_value * (1 + fund_em_info_df.iloc[i]['日增长率'] / 100)) - red_net_value_list.append(last_red_net_value) - - # print('date:',fund_em_info_df.iloc[i]['净值日期'], '单位净值:', fund_em_info_df.iloc[i]['单位净值'], \ - # '日增长率:', fund_em_info_df.iloc[i]['日增长率'], '累计净值', fund_em_info_df.iloc[i]['累计净值']) - except: - print(fund_id, '日增长率数据异常:', fund_em_info_df.iloc[i]['日增长率']) - return '' - else: - try: - last_red_net_value = float(fund_em_info_df.iloc[i]['累计净值']) - red_net_value_list.append(last_red_net_value) - except: - print(fund_id, '累计净值数据异常:', fund_em_info_df.iloc[i]['累计净值']) - return '' - fund_em_info_df['复权净值'] = red_net_value_list - - # print(fund_em_info_df) - fund_em_info_df.to_csv(filename, encoding='utf_8_sig',index=None) - else: - fund_em_info_df = pd.read_csv(filename, dtype=object) - fund_em_info_df.set_index(fund_em_info_df['净值日期']) - except Exception as e: - print(f"错误信息：{e}") - - return fund_em_info_df - -# 绘制基金净值走势 -def draw_cumulative_net_value_trend(x, y, y_hushen300): - # plt.style.use('dark_background') - plt.rcParams["font.sans-serif"]=["SimHei"] #设置字体 - plt.rcParams["axes.unicode_minus"]=False #该语句解决图像中的“-”负号的乱码问题 - - fig, ax = plt.subplots() - # fig.subplots_adjust(bottom=0.2) - - # y_major_locator = MultipleLocator(0.1) - - # # 设置X轴刻度为日期时间 - # ax.xaxis.set_major_formatter(mdate.DateFormatter('%Y-%m-%d')) # 设置时间标签显示格式 - - # ax.yaxis.set_major_locator(y_major_locator) - - # ax.xaxis_date() - # X轴刻度文字倾斜45度 - # plt.xticks(rotation=45) - ax.set_title("基金走势图") - ax.set_xlabel("时间") - ax.set_ylabel("基金累计净值") - ax.plot(x, y, color='r', label="自选组合") - ax.plot(x, y_hushen300, color='b', label="沪深300-110020") - ax.legend() - plt.show() - -# 计算基金成立时间 -def calc_fund_established_time(df): - # print(df.dtypes) - # print(df.iloc[-1, 0], df.iloc[0, 0], df.iloc[-1, 0]-df.iloc[0, 0]) - established_time = (df.iloc[-1, 0] - df.iloc[0, 0]).days/365 - return established_time - -# 计算基金最大回撤 -def calc_fund_drawdown(fund_df, start='1970-01-01', end='2200-01-01'): - df = fund_df[['净值日期', '复权净值']].copy() - df['净值日期'] = pd.to_datetime(df['净值日期']) - - st = df['净值日期'] >= start - et = df['净值日期'] <= end - res = st == et - df = df[res] - - df['max2here'] = df['复权净值'].expanding().max() - df['dd2here'] = df['复权净值']/df['max2here'] - - end_date, remains = tuple(df.sort_values(by=['dd2here']).iloc[0][['净值日期', 'dd2here']]) - - # 计算最大回撤开始时间 - start_date = df[df['净值日期']<=end_date].sort_values(by='复权净值', ascending=False).iloc[0]['净值日期'] - - drawdown = round((1-remains)*100, 2) - print('最大回撤 (%):', drawdown) - print('最大回撤开始时间:', start_date) - print('最大回撤结束时间:', end_date) - - return drawdown, start_date, end_date - -# 计算基金年华回报率 -def cal_fund_annual_return(fund_df, start='1970-01-01', end='2200-01-01'): - df = fund_df[['净值日期', '复权净值']].copy() - # df['净值日期'] = pd.to_datetime(df['净值日期']) - # - # df['日收益'] = df['累计净值'] - df['累计净值'].shift(1) - # df['日收益率'] = (df['累计净值'] - df['累计净值'].shift(1))/df['累计净值'].shift(1) - - st = df['净值日期'] >= start - et = df['净值日期'] <= end - res = st == et - df = df[res] - # df = df[df['净值日期'] >= '2020-01-09'] - - # 计算累计收益率 - # df['累计收益率'] = df['日收益率'].expanding().sum() - total_return = round(((df.iloc[-1, 1] / df.iloc[0, 1]) - 1)*100, 2) - # df = df.fillna(0.0) - # print(df) - - years = round(calc_fund_established_time(df), 2) - # print('分析周期:%f years'%years) - - # 计算 - # print("最新累计收益率:%f, "%df.iloc[-1, 4] + '初始累计收益率:%f'%df.iloc[0, 4]) - if years >= 1: - annualized_returns = (df.iloc[-1, 1] / df.iloc[0, 1])**(1/years) - 1 - else: - annualized_returns = (df.iloc[-1, 1] / df.iloc[0, 1]) - 1 - - annualized_returns = annualized_returns*100 - - print('annualized_returns:%f'%annualized_returns + '%') - - return annualized_returns, total_return; - -# 计算基金夏普比率 -def cal_fund_sharpe_ratio(fund_df, withdrawal, start='1970-01-01', end='2200-01-01'): - df = fund_df[['净值日期', '日增长率']].copy() - - st = df['净值日期'] >= start - et = df['净值日期'] <= end - res = st == et - df = df[res] - # print(df) - - # 计算日收益率的均值 - daily_return_ratio_average = np.mean(df['日增长率'].astype(float))*100 - # print('daily_return_ratio_average:%f'%daily_return_ratio_average) - - # 计算收益率的标准方差 - return_ratio_std = df['日增长率'].astype(float).std() - - # 计算无风险日收益率 - daily_risk_free_return_ratio = (((1 + risk_free_annual_return_ratio)**(1/365)) - 1) * 100 - # print('daily_risk_free_return_ratio:%f' % daily_risk_free_return_ratio) - - # 计算夏普比率 - try: - fund_sharpe_ratio = (daily_return_ratio_average - daily_risk_free_return_ratio) / return_ratio_std * math.sqrt(trading_days_per_year) / 100 - except: - fund_sharpe_ratio = 0.0 - - # 计算卡玛比率 - try: - fund_calmar_ratio = (daily_return_ratio_average - daily_risk_free_return_ratio) / withdrawal * 100 - except: - fund_calmar_ratio = 0.0 - - print('sharpe_ratio:%f' % fund_sharpe_ratio + ', calmar_ratio:%f' % fund_calmar_ratio) - return fund_sharpe_ratio, fund_calmar_ratio - -# 计算基金波动 -def calc_fund_volatility(df_fund, start='1970-01-01', end='2200-01-01'): - # difflntotal = [] - - df = df_fund[['净值日期', '复权净值', '上一交易日复权净值']].copy() - # df['上一次累计净值'] = df['累计净值'].shift(1) - - st = df['净值日期'] >= start - et = df['净值日期'] <= end - res = st == et - df = df[res] - - # days = df.shape[0] - # print(days) - - df['价格自然对数差'] = np.log(df['复权净值']) - np.log(df['上一交易日复权净值']) - volitality = np.std(df['价格自然对数差']) * 100 * math.sqrt(trading_days_per_year) - - # 与前一种计算结果一致 - # df['价格比率对数'] = np.log(df['累计净值'] / df['上一次累计净值']) - # volitality = np.std(df['价格比率对数'])* 100 * math.sqrt(trading_days_per_year) - - print('volitality:%f'%volitality) - - return volitality - -# 计算基金kpi -def calc_fund_kpi(fund_id, df_cumulative_net_value_trend, start='1970-01-01', end='2200-01-01'): - dict_select = { - 'code': fund_id, # 基金代码 - 'years': '', # 成立时间 - 'withdrawal': '', # 最大回撤 - 'annual_return': '', # 年化收益率 - 'total_return': '', # 累计收益率 - 'sharp': '', # 夏普比率 - 'calmar': '', # 卡玛比率 - 'volatility': '', # 波动率 - } - - df = df_cumulative_net_value_trend[['净值日期', '复权净值', '日增长率']].copy() - - # 生成上一个交易日累计净值 - df['上一交易日复权净值'] = df['复权净值'].shift(1) - - # 数据格式转换 - df['净值日期'] = pd.to_datetime(df['净值日期']) - df['复权净值'] = df['复权净值'].astype(float) - df['上一交易日复权净值'] = df['上一交易日复权净值'].astype(float) - - # # 计算日收益和日收益率 - # df['日收益'] = df['累计净值'] - df['上一交易日累计净值'] - # df['日收益率'] = (df['累计净值'] - df['上一交易日累计净值'])/df['上一交易日累计净值'] - - # x = df['净值日期'] - # y = df['累计净值'] - # - # # 绘制基金累计净值走势图 - # draw_cumulative_net_value_trend(x, y) - - # 计算成立时间 - years = calc_fund_established_time(df) - dict_select['years'] = round(years, 2) - print('基金[%s'%fund_id + ']成立时间:%d'%(int(years)) + '年%d'%((years-int(years))*365) + '天') - - # 计算最大回撤 - withdrawal, start_date, end_date = calc_fund_drawdown(df, start, end) - dict_select['withdrawal'] = round(withdrawal, 2) - - # 计算年化收益 - annual_return, total_return = cal_fund_annual_return(df, start, end) - dict_select['annual_return'] = round(annual_return, 2) - dict_select['total_return'] = round(total_return, 2) - - # 计算夏普比率 - sharpe_ratio, calmar_ratio = cal_fund_sharpe_ratio(df, withdrawal, start, end) - dict_select['sharp'] = round(sharpe_ratio, 2) - dict_select['calmar'] = round(calmar_ratio, 4) - - # 计算波动率 - volatility = calc_fund_volatility(df, start, end) - dict_select['volatility'] = round(volatility, 2) - - print(dict_select) - - return dict_select - -# user_agent列表 -user_agent_list = [ - 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/21.0.1180.71 Safari/537.1 LBBROWSER', - 'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; QQDownload 732; .NET4.0C; .NET4.0E)', - 'Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.84 Safari/535.11 SE 2.X MetaSr 1.0', - 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Maxthon/4.4.3.4000 Chrome/30.0.1599.101 Safari/537.36', - 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.122 UBrowser/4.0.3214.0 Safari/537.36' -] - -# referer列表 -referer_list = [ - 'http://fund.eastmoney.com/110022.html', - 'http://fund.eastmoney.com/110023.html', - 'http://fund.eastmoney.com/', - 'http://fund.eastmoney.com/110025.html' -] - -# 爬取网页基金数据 -def get_url(baseUrl): - # 获取一个随机user_agent和Referer - headers = {'User-Agent': random.choice(user_agent_list), 'Referer': random.choice(referer_list)} - try: - resp = requests.get(baseUrl, headers=headers) - # print(resp.status_code) - if resp.status_code == 200: - # print(resp.content) - return resp.text - print("没有爬取到相应的内容") - return None - except: - print("没有爬取到相应的内容") - return None - -# 获得基金费用 -def get_fund_fee(code): - dict_select = { - '基金代码': code, # 基金代码 - 'scale': '', # 基金规模 - 'm_fee': '', # 管理费 - 'c_fee': '', # 托管费 - 'sale_fee': '', # 销售费用 - 'sub_fee': '', # 申购费<50W - # 'buy_fee_less_than_100': '', # 申购费<100W - # 'buy_fee_more_than_100': '', # 申购费>=100W - } - - url = 'http://fundf10.eastmoney.com/jjfl_%s' % code + '.html' - html = get_url(url) - soup = BeautifulSoup(html, 'html.parser') - # log(soup.prettify()) - - target = '' - for co in soup.find_all(['span', 'td', 'span', 'label']): - # log('target%s' % target) - log('co.text:%s' % co.text) - if target != '': - # '''dict_select[target] = co.text.encode('utf-8')''' - dict_select[target] = co.text - if target == 'sale_fee': - break - target = '' - - if co.text.find(u'资产规模') >= 0: - target = 'scale' - elif co.text == u'管理费率': - target = 'm_fee' - elif co.text == u'托管费率': - target = 'c_fee' - elif co.text == u'销售服务费率': - target = 'sale_fee' - - # print(dict_select) - - try: - dict_select['scale'] = float(get_decimal_suffix(dict_select['scale'], '亿元')) - if dict_select['scale'] == '': - dict_select['scale'] = 0.0 - except: - dict_select['scale'] = 0.0 - - try: - dict_select['m_fee'] = float(get_decimal(dict_select['m_fee'])) - if dict_select['m_fee'] == '': - dict_select['m_fee'] = 0.0 - except: - dict_select['m_fee'] = 0.0 - - try: - dict_select['c_fee'] = float(get_decimal(dict_select['c_fee'])) - if dict_select['c_fee'] == '': - dict_select['c_fee'] = 0.0 - except: - dict_select['c_fee'] = 0.0 - - try: - dict_select['sale_fee'] = float(get_decimal(dict_select['sale_fee'])) - if dict_select['sale_fee'] == '': - dict_select['sale_fee'] = 0.0 - except: - dict_select['sale_fee'] = 0.0 - - print(dict_select) - return dict_select - -# 获得基金费用列表 -def get_fund_fee_list(df): - err_count = 0 - fee_list = [] - filename = 'fund_fee_list.csv' - filename_bak = 'fund_fee_list-bak.csv' - # print(filename) - - if not os.path.exists(filename): - for i in np.arange(0, df.shape[0]): - # try: - print('基金编码[%s'%(df.iloc[i]['基金代码']) + ']正在读取费用...') - dict_item = get_fund_fee(df.iloc[i]['基金代码']) - dict_item['sub_fee'] = df.iloc[i]['手续费'] - fee_list.append(dict_item) - # except: - # err_count = err_count + 1 - # print('基金编码[%s' % (df.iloc[i]['基金代码']) + ']读取费用失败...') - time.sleep(1) - - df_fee = pd.DataFrame(fee_list) - df_fee.to_csv(filename, encoding='utf_8_sig', index=None) - print('读取费率失败次数:%d' % err_count) - else: - df_fee = pd.read_csv(filename, dtype=object) - # for i in np.arange(0, df.shape[0]): - # try: - # print(df_fee.iloc[i]['基金代码'], df_fee.iloc[i]['sub_fee']) - # df_fee.iloc[i]['sub_fee'] = get_decimal(df_fee.iloc[i]['sub_fee']) - # except: - # df_fee.iloc[i]['sub_fee'] = 0.0 - # - # df_fee.to_csv(filename_bak, encoding='utf_8_sig', index=None) - - return df_fee - -# 计算基金rank -def get_fund_rank(df_in, keywords, max_withdrawal, establish_year=3.0, start='1970-01-01', end='2200-01-01'): - fund_kpi_list = [] - - # 根据关键字检索 - df_dist = query_fund_by_fundname_keyword(df_in, keywords) - - df_dist = df_dist[df_dist['基金类型'].str.contains('场内') == False] - for i in np.arange(0, df_dist.shape[0]): - try: - print('基金编码[%s'%(df_dist.iloc[i]['基金代码']) + ']正在读取...') - - # 爬取历史净值 - df_cumulative_net_value_trend = get_fund_his_by_id(df_dist.iloc[i]['基金代码'], dir_cumulative_net_value_trend) - print('历史数据已加载...') - if (keywords != '债') or (keywords == '债' and df_cumulative_net_value_trend['日增长率'].astype(float).max(skipna=True) <= 2.0): - # 计算基金KPI - dict_kpi = calc_fund_kpi(df_dist.iloc[i]['基金代码'], df_cumulative_net_value_trend, start, end) - dict_kpi['code'] = df_dist.iloc[i]['基金代码'] - dict_kpi['name'] = df_dist.iloc[i]['基金简称'] - dict_kpi['type'] = df_dist.iloc[i]['基金类型'] - print('KPI计算已完成...') - - # 爬取费用 - dict_kpi['scale'] = df_dist.iloc[i]['scale'] - dict_kpi['m_fee'] = df_dist.iloc[i]['m_fee'] - dict_kpi['c_fee'] = df_dist.iloc[i]['c_fee'] - dict_kpi['sale_fee'] = df_dist.iloc[i]['sale_fee'] - dict_kpi['sub_fee'] = df_dist.iloc[i]['sub_fee'] - - # print(dict_kpi) - fund_kpi_list.append(dict_kpi) - # time.sleep(1) - except Exception as e: # 未知异常的捕获 - print(f"异常信息:{e}") - print('基金编码[%s'%(df_dist.iloc[i]['基金代码']) + ']读取失败') - # # time.sleep(1) - - # print(fund_kpi_list) - df_kpi = pd.DataFrame(fund_kpi_list) - print(df_kpi) - - # 删除异常数据行 - # df_kpi.dropna(subset=['withdrawal'], inplace=True) - - # 对数据筛选 - df_kpi = df_kpi[df_kpi['years'].astype(float) >= establish_year] - df_kpi = df_kpi[df_kpi['withdrawal'].astype(float) <= max_withdrawal] - - try: - df_kpi['total_fee'] = df_kpi['sub_fee'].astype(float) + df_kpi['m_fee'].astype(float) + df_kpi['c_fee'].astype(float) + df_kpi['sale_fee'].astype(float) - except Exception as e: # 未知异常的捕获 - print(f"异常信息:{e}") - df_kpi['total_fee'] = 0 - print('[%s'%keywords + ']计算累计费用失败，跳过!') - - # 对数据排序, 'sharp', 'total_fee', 'annual_return' - df_kpi.sort_values(by=['sharp', 'withdrawal', 'calmar', 'total_fee'], ascending=[False, False, False, True], inplace=True) - - # print(df_kpi.dtypes) - # df_kpi.to_csv('fund_kpi.csv') - df_kpi.to_excel('fund_rank-%s' % keywords + '.xlsx') - print(df_kpi) - return df_kpi - -# 计算基金组合预测数据 -def calc_fund_comb_kpi(fund_comb, start='1970-01-01', end='2200-01-01'): - df_comb = pd.DataFrame(columns=['净值日期', '复权净值', '复权净值临时', '日增长率']) - - for i in np.arange(0, fund_comb.shape[0]): - # 爬取历史净值 - df = get_fund_his_by_id(fund_comb.iloc[i]['code'], dir_cumulative_net_value_trend) - - # df.dropna(axis=0, inplace=True) - # df.fillna(method='pad',axis=0, inplace=True) - - # 过滤时间 - st = df['净值日期'] >= start - et = df['净值日期'] <= end - res = st == et - df = df[res] - - if df_comb.empty: - df_comb['净值日期'] = df['净值日期'] - df_comb.set_index('净值日期', inplace=True) - - df.set_index('净值日期', inplace=True) - # print(df) - - print(fund_comb.iloc[i]['share']) - # if pd.isnull(df_comb.iloc[0]['复权净值']): - if df_comb['复权净值'].isnull().all(): - df_comb['复权净值临时'] = 0.0 - else: - df_comb.loc[:, '复权净值临时'] = df_comb.loc[:, '复权净值'] - - df_comb.loc[:, '复权净值'] = df.loc[:, '复权净值'].astype(float) * fund_comb.iloc[i]['share'] - # print(type(df_comb.loc[:, '复权净值']), type(df_comb.loc[:, '复权净值临时'])) - df_comb.loc[:, '复权净值'] = df_comb.loc[:, '复权净值'].astype(float).add(df_comb.loc[:, '复权净值临时']) - df_comb.fillna(method='pad', axis=0, inplace=True) - - # 计算日增长率 - df_comb['日增长率'] = (df_comb['复权净值'] - df_comb['复权净值'].shift(1))/df_comb['复权净值'].shift(1) - - # print(df_comb) - return df_comb - -if __name__ == "__main__": - pd.set_option('display.max_rows', 1000) - pd.set_option('display.max_columns', 10) - - # get_fund_his_by_id('000051', dir_cumulative_net_value_trend) - - # 获得公募基金列表 - df = get_all_fund_outline() - # df = get_all_open_fund_daily() - df = df[df['申购状态'] == '开放申购'] - - # 加载费用 - df_fee = get_fund_fee_list(df) - df_fee = df_fee[['基金代码', 'scale', 'm_fee', 'c_fee', 'sale_fee', 'sub_fee']].copy() - # get_fund_fee('000051') - - df = pd.merge(df, df_fee, on='基金代码') - df.set_index(df['基金代码']) - - # 分红测试代码 - # df_dividend = get_fund_dividend_by_id('002898') - # df_dividend = query_last_dividend_before_date(dir_cumulative_net_value_trend, '002898', '2021-01-15') - # print(type(df_dividend)) - # print(df_dividend) - - df_kpi_csi300 = get_fund_rank(df, '沪深300', 60.0, 5, '2018-01-01', '2022-10-30') - df_kpi_csi500 = get_fund_rank(df, '中证500', 50.0, 5, '2018-01-01', '2022-10-30') - df_kpi_gem = get_fund_rank(df, '创业板', 50.0, 5, '2018-01-01', '2022-10-30') - df_kpi_gold = get_fund_rank(df, '黄金', 50.0, 5, '2018-01-01', '2022-10-30') - df_kpi_bond = get_fund_rank(df, '债', 30.0, 5, '2018-01-01', '2022-10-30') - df_kpi_sp500 = get_fund_rank(df, '标普500', 50.0, 5, '2018-01-01', '2022-10-30') - df_kpi_nasda = get_fund_rank(df, '纳斯达克', 50.0, 5, '2018-01-01', '2022-10-30') - # - # # 行业基金 - # get_fund_rank(df, '新能源', 50.0, 3, '2018-01-01', '2021-02-18') - # get_fund_rank(df, '消费', 50.0, 3, '2018-01-01', '2021-02-18') - # get_fund_rank(df, '商品', 50.0, 3, '2018-01-01', '2021-02-18') - # get_fund_rank(df, '混合', 50.0, 3, '2018-01-01', '2021-03-28') - - print(f"基金组合:{df_kpi_csi300.iloc[0]['code']}:0.25,\n \ - {df_kpi_csi500.iloc[0]['code']}:0.15,\n \ - {df_kpi_gem.iloc[0]['code']}: 0.10,\n \ - {df_kpi_gold.iloc[0]['code']}: 0.10,\n \ - {df_kpi_bond.iloc[0]['code']}: 0.20],\n \ - {df_kpi_sp500.iloc[0]['code']}: 0.10],\n\ - {df_kpi_nasda.iloc[0]['code']}: 0.10]") - # # 组合数据 - # fund_comb = pd.DataFrame([['673100', 0.50], ['001879', 0.15], ['000187', 0.20], ['000218', 0.15]], columns=['code', 'share']) - fund_comb = pd.DataFrame([ - [df_kpi_csi300.iloc[0]['code'], 0.25], - [df_kpi_csi500.iloc[0]['code'], 0.15], - [df_kpi_gem.iloc[0]['code'], 0.10], - [df_kpi_gold.iloc[0]['code'], 0.10], - [df_kpi_bond.iloc[0]['code'], 0.20], - [df_kpi_sp500.iloc[0]['code'], 0.10], - [df_kpi_nasda.iloc[0]['code'], 0.10] - ], - columns=['code', 'share']) - df_comb = calc_fund_comb_kpi(fund_comb, '2017-06-02', '2022-10-30') - df_comb.reset_index(inplace=True) - x = df_comb['净值日期'] - y = df_comb['复权净值'] - - fund_hs300 = pd.DataFrame([['000512', 1.00]], columns=['code', 'share']) - df_hs300 = calc_fund_comb_kpi(fund_hs300, '2017-06-02', '2022-10-30') - df_hs300.reset_index(inplace=True) - df_hs300.to_csv("参考沪深300.csv", encoding='utf_8_sig', index=None) - - y_hs300 = df_hs300['复权净值'] - - # 绘制基金累计净值走势图 - draw_cumulative_net_value_trend(x, y, y_hs300) diff --git a/start.sh b/start.sh new file mode 100755 index 0000000..1634666 --- /dev/null +++ b/start.sh @@ -0,0 +1,7 @@ +cd demo-widget + +# Create tables & set menu +yao migrate --reset +yao run flows.setmenu + +yao start -- GitLab