kawine/dataset_difficulty

Issues

• Are there any differences between finetuning process on null input and label and directly calculating distribution of labels to obtain probabilities?

  #10 opened 6 months ago by isaacveg
  0

• Why use the the same dataset for the train and validation data ?

  #7 opened a year ago by cfsmile
  2

• Can PVI be used to measure usable information for non classification tasks, i.e., data reconstruction, object detection, etc,.

  #5 opened a year ago by Zmjcc
  1

• can we use V-information to rank the difficulty of training instances?

  #4 opened a year ago by Indiiigo
  1

• Error with python augment.py --raw_data_dir RAW_DATA_DIR

  #3 opened 2 years ago by khareyash05
  1

• Error missing argument "train_size"

  #2 opened 2 years ago by khareyash05
  2

• Requirements.txt

  #1 opened 2 years ago by venelink
  1