dataframe先分组运算再合并输出:主要用到分组函数groupby和合并函数append、concat
,具体代码:
比如先分组,对每组数据进行删除异常值处理(MAD),然后将处理后的数据合并成一个dataframe输出。
import os
import pandas as pd
from pyod.models.mad import MAD
import warnings
warnings.filterwarnings('ignore')
#%%
#test
df = pd.read_csv('./21.csv')
# print(len(df))#2273824
def fit_model(model, data, column='s'):
# fit the model and predict it
df = data.copy()
data_to_predict = data[column].to_numpy().reshape(-1, 1)
predictions = model.fit_predict(data_to_predict)
df['Predictions_1'] = predictions
return df
df_groupby = df.groupby(["class"])#group by class
frames = []
for decyear,group in df_groupby:
# print(decyear)
# print(group)
mad_model = MAD()
group_mad = fit_model(mad_model,group)
group_mad_filter = group_mad[(group_mad['Predictions_1'] == 0)]
# print(group_mad_filter)
frames.append(group_mad_filter)
result = pd.concat(frames, axis=0, names=None)
print('done!')