def tushare_clean(csv_path, override=True, pickle_name=None):
"""
1. save to local csv
2. save to local pickle
"""
def clean(df):
df.reset_index(drop=True, inplace=True)
df['date'] = pd.DatetimeIndex(df['date'])
df.set_index('date', inplace=True)
return df
walk_list = os.walk(csv_path).next()
csv_list=[]
pickle_dict ={}
for i in walk_list[2]:
if 'csv' in i:
df = pd.read_csv(os.path.join(csv_path, '%s' % i),
parse_dates=True,index_col=0)
cleaned_df = clean(df)
# override CSV
if override:
cleaned_df.to_csv(os.path.join(csv_path, '%s' % i))
# create pickle
if type(pickle_name) is str:
symbol = i.replace('.csv','')
pickle_dict[symbol] = cleaned_df
# Save to pickle
if type(pickle_name) is str:
pd.to_pickle(pickle_dict, os.path.join(csv_path, '%s.pkl' % pickle_name))
评论列表
文章目录