libffcv/ffcv-imagenet

Issues

• Error in val_loop

  #22 opened 10 months ago by AndreaBrg
  0

• numpy.core._exceptions._ArrayMemoryError: Unable to allocate 2.73 TiB for an array with shape (124994103978,) and data type [('sample_id', '<u8'), ('ptr', '<u8'), ('size', '<u8’)]

  #21 opened a year ago by TheDarkKnight-21th
  0

• ValueError: could not broadcast input array from shape (80,) into shape (160,)

  #18 opened 2 years ago by FrancescoSaverioZuppichini
  1

• yaml file for imagenet writter hyperparameters?

  #17 opened 2 years ago by realliyifei
  0

• How to enable Multi-GPU training (1 model, multiple GPUs) under the server with limited memory?

  #10 opened 3 years ago by fantasysee
  3

• train_loss variable is None and val_loss variable in log not showing up

  #5 opened 3 years ago by rraju1
  5

• ImportError: libopencv_imgproc.so

  #7 opened 3 years ago by eldarkurtic
  4

• Potential minor speedup: Gaussian blur is a separable 2d convolution

  #16 opened 3 years ago by lebrice
  1

• DEFAULT_CROP_RATIO wrong?

  #9 opened 3 years ago by ByungKwanLee
  1

• batch_size=1 causes error when Squeeze() is in the "label" pipeline

  #13 opened 3 years ago by botcs
  1

• ValueError(“total size of new array must be changed”)

  #12 opened 3 years ago by cindy-cheng0214
  0

• Training extremely slow

  #11 opened 3 years ago by netw0rkf10w
  4

• Question about the parameter of the `write_imagenet.py`

  #8 opened 3 years ago by mzhaoshuai
  2

• Imagenet dataset preparation size

  #1 opened 3 years ago by aniketrege
  8

• PR for resuming training from checkpoint and validate model loaded from disk

  #6 opened 3 years ago by aniketrege
  0

• Reproducing Validation Numbers

  #4 opened 3 years ago by aniketrege
  1

• for training imagenet, how to train in multiple machines with same gpu numbers?

  #3 opened 3 years ago by jeannotes
  1

• A complete example for imagenet data loading

  #2 opened 3 years ago by gd-zhang
  1