更改列名称.rename()
# 更改列名称.rename()
df1.rename(columns={'Ta_10m': 'Ta'}, inplace=True)
将时间改成时间戳pd.to_datetime()
df1["TIMESTAMP"] = pd.to_datetime(df1["TIMESTAMP"], errors='ignore')
时间取整.dt.round()
# 这里是让一些不整点的时间整起来
freq = '1min'
df1["Date"] = df1['Date'].dt.round(freq)
舍弃时分秒将时间提取至年月日.dt.date
df1['date'] = df1['Date'].dt.date
将时间从年月日转为一年中的第几天(DOY,day of the year).dt.dayofyear
df1['DOY'] = df1['Date'].dt.dayofyear
删除列.drop()
df1.drop(labels = ["date"], axis = 1, inplace = True) # 删除列,需要增加inplace = True这一项
提取所有列名称并转为列表.columns.to_list()
header_list = df1.columns.to_list()
筛选并保留满足条件的行.loc()
# 筛选出LE≠-6999且H≠-6999的行,保存为df1
df1 = df.loc[(data['LE'] != -6999) | (data['Hs'] != -6999)]
拆分路径或名称.split()
file = "D:\\YMJ_file\\1_last_plz\\fluxnet\\test.csv"
outpath = file.split(".csv")[0] + ".xlsx"
# "D:\\YMJ_file\\1_last_plz\\fluxnet\\test.xlsx"
缺失值填充.fillna()、数据替换.replace()
# 将所有空值替换成-6999
data.fillna(-6999, inplace=True)
# 将所有-9999的值替换成-6999
data.replace(-9999, -6999, inplace=True)