我需要比较一个名为 last_login 的列,如果日期与今天的日期匹配,我想将整行检索到列表或类似的东西:
def joined_today(self, df):
users_joined_today = []
date_joined = pd.DataFrame(df)
today = datetime.date.today()
for i in date_joined['last login']:
i = i.date()
if i == today:
users_joined_today.append(i)
return users_joined_today
我只是想知道什么是检索与 joined_today() 函数返回的值匹配的整行的有效方法?
请您参考如下方法:
对于 Pandas,您应该致力于使用矢量化操作:
# convert series to datetime, if not already
df['last_login'] = pd.to_datetime(df['last_login'])
# calculate Boolean series mask
mask = df['last_login'].dt.normalize() == pd.to_datetime('today')
# apply mask
df_filtered = df[mask]
# optionally, convert to list of lists
df_filtered_L = df_filtered.values.tolist()
规范化 datetime series 将时间分量展平为零,因此您可以将其与 pd.to_datetime('today') 进行比较,这已经归一化了。
例如,pd.to_datetime('now').normalize() == pd.to_datetime('today')返回 True .
