search

google-research / mixmatch

Apache License 2.0
1.13k stars 163 forks source link

What is the purpose of creating the "-map.json" file in scripts/create_split.py and eval_labeled/eval_unlabeled in libml/data.py? #13

Closed varunnair18 closed 5 years ago

varunnair18 commented 5 years ago

Does the "-map.json" file have any purpose in the training process at all? I see that it stores the ids of images that are used as unlabeled data, but cannot find any usage of it anywhere in the repository.

Can you also explain why you read from the "-label.tfrecord" and "-unlabel.tfrecord" twice (for train_labeled, train_unlabeled, eval_labeled, eval_unlabeled respectively) in libml/data.py? Is this because you fetch a sample of images for training and then later fetch the same set of samples to compute evaluation metrics?

david-berthelot commented 5 years ago

JSON is just for information, it's not used. For the second question you guessed right.

varunnair18 commented 5 years ago

Good to know, thank you for helping!