Problem in training unet


  1. Update to the latest docker which is released yesterday. See Migrating to TAO Toolkit — TAO Toolkit 3.0 documentation and NVIDIA NGC
  2. Please use loss: "cross_dice_sum"
  3. Please convert the mask images to gray images. After checking the public dataset you mentioned, the pixel value is either 0 or 128. Please map 128 to 1.
  4. For binary segmentation, the label_id should be 0 and 1. BTW, if there are 4 classes , label_id should belong to 0-3
1 Like

Hi @Morganh
Thank you
I will try.