dasoto / skincancer Goto Github PK

root@dos-ubuntu-s-8vcpu-16gb-nyc1-01:~/hub-skin-cancer# pip install -r requirements.txt
Collecting h5py<3.0.0
  Downloading h5py-2.10.0-cp38-cp38-manylinux1_x86_64.whl (2.9 MB)
     |████████████████████████████████| 2.9 MB 18.1 MB/s
ERROR: Could not find a version that satisfies the requirement tensorflow==1.15 (from -r requirements.txt (line 2)) (from versions: 2.2.0rc1, 2.2.0rc2, 2.2.0rc3, 2.2.0rc4, 2.2.0, 2.2.1, 2.2.2, 2.3.0rc0, 2.3.0rc1, 2.3.0rc2, 2.3.0, 2.3.1, 2.3.2, 2.4.0rc0, 2.4.0rc1, 2.4.0rc2, 2.4.0rc3, 2.4.0rc4, 2.4.0, 2.4.1, 2.5.0rc0, 2.5.0rc1, 2.5.0rc2, 2.5.0rc3, 2.5.0)
ERROR: No matching distribution found for tensorflow==1.15 (from -r requirements.txt (line 2))

ubuntu 20.04 from scratch

Hey, first of all great project and very informative. As you have described all the models that you have trained, there is no description of mymodel-2.h5. Can you provide the description for that. Also the other model trained using VGG16 pretrained models is it available for download?

One more thing, could you describe the evaluation method for the scores obtained after predictions, like what threshold did you keep for classification.

Thank you.

When I tried to train with your training script (train_model. new. py), I found that ACC curve and loss performed very badly. What might be the reason? Is my way of data preprocessing wrong?
Thank you very much for any help.

Please specify necessary steps to run this project.