|
|
@@ -80,8 +80,6 @@ def segment_calc_diff_rate_by_score(df: pd.DataFrame, segment_file_path: str, st
|
|
|
segment_label_sum=('label', 'sum'),
|
|
|
segment_label_cnt=('label', 'count'),
|
|
|
segment_score_avg=('score', 'mean'),
|
|
|
- p_cpm_avg=('p_cpm', 'mean'),
|
|
|
- t_cpm_avg=('t_cpm', 'mean'),
|
|
|
).reset_index()
|
|
|
group_df['segment_true_score'] = group_df['segment_label_sum'] / group_df['segment_label_cnt']
|
|
|
group_df['segment_diff_rate'] = (group_df['segment_score_avg'] / group_df['segment_true_score'] - 1).mask(group_df['segment_true_score'] == 0, 0)
|
|
|
@@ -156,9 +154,9 @@ def _main(old_predict_path: str, new_predict_path: str, calibration_file: str, a
|
|
|
# 按照曝光排序，写入本地文件
|
|
|
merged = merged.sort_values(by=['view'], ascending=False)
|
|
|
merged = merged[[
|
|
|
- 'cid', 'view', "conv", "true_ctcvr", 'true_cpm',
|
|
|
- "old_score_avg", "new_score_avg", "(old-true)/true", "(new-true)/true", "old_cpm", "new_cpm",
|
|
|
- "old_score_2_avg", "new_score_2_avg", "(old2-true)/true", "(new2-true)/true", "old2_cpm", "new2_cpm",
|
|
|
+ 'cid', 'view', "conv", "true_ctcvr",
|
|
|
+ "old_score_avg", "new_score_avg", "(old-true)/true", "(new-true)/true",
|
|
|
+ "old_score_2_avg", "new_score_2_avg", "(old2-true)/true", "(new2-true)/true",
|
|
|
]]
|
|
|
|
|
|
# 根据文件名保存不同的格式
|
