data-to-loacl

main_spider.py

@@ -76,7 +76,7 @@ class LightGBM(object):
             features[key] = pd.to_numeric(features[key], errors="coerce")
         for key in self.str_columns:
             features[key] = self.label_encoder.fit_transform(features[key])
-            # self.label_mapping[key] = dict(zip(self.label_encoder.classes_, self.label_encoder.transform(self.label_encoder.classes_)))
+            self.label_mapping[key] = dict(zip(self.label_encoder.classes_, self.label_encoder.transform(self.label_encoder.classes_)))
         return features, labels, video_ids, video_titles
 
     def best_params(self):
@@ -187,7 +187,7 @@ class LightGBM(object):
         # 水平合并
         df_concatenated = pd.concat([ids, titles, x, true_label_df, pred_score_df, pred_label_df], axis=1)
         for key in self.str_columns:
-            df_concatenated[key] = self.label_encoder.inverse_transform(df_concatenated[key])
+            df_concatenated[key] = self.label_mapping[key][df_concatenated[key]]
         df_concatenated.to_excel("data/predict_data/spider_predict_result_{}.xlsx".format(dt), index=False)
 
     def feature_importance(self):