import tushare as ts
import pandas as pd
import time
import os
import datetime
# 指定自己要存放文件的絕對路徑
os.chdir('E:/')
pd.set_option('expand_frame_repr', False)
now_time = datetime.date.today()
# 從tushare獲取指定日期
def get_today_all_ts(date):date_now = date# 先獲得所有股票的收盤數據df_close = ts.get_today_all()# 獲得所有股票的基本信息df_basics = ts.get_stock_basics()df_all = pd.merge(left=df_close, right=df_basics, on='name', how='outer')df_all['code'] = df_all['code'].astype(str) + ' '# 保存數據df_all.to_csv(str(date_now) + '_ts.csv', index=False, encoding='gbk')print('%s is downloaded.' % (str(date_now)))print(df_all)return df_allif __name__ == '__main__':get_today_all_ts(date=now_time)

import pandas as pd
import requests
import tushare as ts
import os
import time
import warnings
warnings.filterwarnings('ignore')
pd.set_option('expand_frame_repr', False)
os.chdir('E:/')
date_now = datetime.date.today()
print(str(date_now))
# ====龍虎榜數據# 讀取TDX的每日股票數據
df = pd.read_csv('{}_ts.csv'.format(str(date_now)), encoding='gbk')
df.fillna(0, inplace=True)
df.replace('nan ', 0, inplace=True)
df['timeToMarket'] = pd.to_datetime(df['timeToMarket'])df[['changepercent', 'pe', 'mktcap', 'nmc']] = df[['changepercent', 'pe', 'mktcap', 'nmc']].astype(float)
df['timeToMarket'] = pd.to_datetime(df['timeToMarket'])df['code'] = df['code'].astype(str)  # 轉化成str后，NAN也變成nan str格式
# 添加 交易所 列
df.loc[df['code'].str.startswith('3'), 'exchange'] = 'CY'
df.loc[df['code'].str.startswith('6'), 'exchange'] = 'SH'
df.loc[df['code'].str.startswith('0'), 'exchange'] = 'SZ'# 找出上漲的股票
df_up = df[df['changepercent'] > 0.00]
# 走平股數
df_even = df[df['changepercent'] == 0.00]
# 找出下跌的股票
df_down = df[df['changepercent'] < 0.00]# 找出漲停的股票
limit_up = df[df['changepercent'] >= 9.70]
limit_down = df[df['changepercent'] <= -9.70]# 漲停股數中的未封板股，上市日期小于15天
limit_up_new = limit_up[pd.to_datetime(date_now) - limit_up['timeToMarket'] <= pd.Timedelta(15)]
# 漲停股數中次新股，上市日期小于1年
limit_up_fresh = limit_up[pd.to_datetime(date_now) - limit_up['timeToMarket'] <= pd.Timedelta(365)]# 漲停股數中的未封板股，上市日期小于15天
limit_down_new = limit_down[pd.to_datetime(date_now) - limit_down['timeToMarket'] <= pd.Timedelta(15)]
# 漲停股數中次新股，上市日期小于1年
limit_down_fresh = limit_down[pd.to_datetime(date_now) - limit_down['timeToMarket'] <= pd.Timedelta(365)]print('A股上漲個數： %d, A股下跌個數： %d, A股走平個數: %d。' % (df_up.shape[0], df_down.shape[0], df_even.shape[0]))
print('A股總成交額：%d, 總成交量：%d' % (df['amount'].sum(), df['volume'].sum()))
print('A股平均市盈率：%.2f， 平均流通市值 %.2f 億, 平均總市值 %.2f 億' % (df['pe'].mean(), df['nmc'].mean(), df['mktcap'].mean()))
print('漲停數量：%d 個, 漲停中上市日期小于15天的：%d, 漲停中上市日期小于1年的：%d' % (limit_up.shape[0], limit_up_new.shape[0], limit_up_fresh.shape[0]))
print('跌停數量：%d 個, 漲停中上市日期小于15天的：%d, 漲停中上市日期小于1年的：%d' % (limit_down.shape[0], limit_down_new.shape[0], limit_down_fresh.shape[0]))# 獲取指定列的分析統計結果
def get_output(df, columns='industry', name='_limit_up'):df = df.copy()output = pd.DataFrame()output = pd.DataFrame(df.groupby(columns)['code'].count())output['pe_mean'] = df.groupby(columns)['pe'].mean()output['pe_median'] = df.groupby(columns)['pe'].median()output['nmc_mean'] = df.groupby(columns)['nmc'].mean()output['nmc_median'] = df.groupby(columns)['nmc'].median()output['mktcap_mean'] = df.groupby(columns)['mktcap'].mean()output['mktcap_median'] = df.groupby(columns)['mktcap'].median()output['volume_mean'] = df.groupby(columns)['volume'].mean()output['volume_median'] = df.groupby(columns)['volume'].median()output['amount_mean'] = df.groupby(columns)['amount'].mean()output['amount_median'] = df.groupby(columns)['amount'].median()output.sort_values('code', ascending=False, inplace=True)output.rename(columns={'code': name + '_count'}, inplace=True)return outputfor i in ['industry', 'exchange', 'area']:
# 對漲停的股票分析output_limit_up = get_output(limit_up, columns=i, name='limit_up')print(output_limit_up)
# 對跌停的股票分析output_limit_down = get_output(limit_down, columns=i, name='limit_down')print(output_limit_down)
# 對全量的股票分析output_total = get_output(df, columns=i, name='total')print(output_total)