search

selimsef / dfdc_deepfake_challenge

A prize winning solution for DFDC challenge
MIT License
779 stars 209 forks source link

Could you explain what does the argument "fold" mean #12

Closed lyak1314 closed 4 years ago

lyak1314 commented 4 years ago

When I run generate_fold.py,I find the "fold" is alway "0"

                for k, v in metadata.items():
                    fold = None
                    for i, fold_dirs in enumerate(folds):
                        #if part in fold_dirs:
                            fold = i
                            break
                    assert fold is not None
                    video_id = k[:-4]
                    video_fold[video_id] = fold

I debug this part and find "fold” is alway "0" and then break Since I can't understand what does "fold" mean,I don't know how to solve it and what is the correct could you explain it in detail? thank you!

selimsef commented 4 years ago

Fold is the index of the fold (cross validation). As I decided to use a holdout for validation istead of kfold split the fold number is always 0

yuzhoufengg2 commented 6 months ago

Fold is the index of the fold (cross validation). As I decided to use a holdout for validation istead of kfold split the fold number is always 0

Fold is always 0,but when I retrain the model,I can not get the train data, due to _prepare_data method.This is a bug？ image image