search

kumar-shridhar / PyTorch-BayesianCNN

Bayesian Convolutional Neural Network with Variational Inference based on Bayes by Backprop in PyTorch.
MIT License
1.42k stars 323 forks source link

Cant get uncertainty estimation for custom dataset #69

Open sharifsagar80 opened 2 years ago

sharifsagar80 commented 2 years ago

I am trying to run uncertainty_estimation file with custom dataset but i am always getting the following error:

using 3306 images for training, 364 images for validation. Traceback (most recent call last): File "uncertainty_estimation.py", line 184, in run(args.net_type, args.weights_path, args.notmnist_dir) File "uncertainty_estimation.py", line 139, in run sample_mnist, truth_mnist = get_sample(mnist_set) File "uncertainty_estimation.py", line 115, in get_sample sample = transform(sample) File "/home/sharif/PyTorch-BayesianCNN-master/venv/lib/python3.8/site-packages/torchvision/transforms/transforms.py", line 60, in call img = t(img) File "/home/sharif/PyTorch-BayesianCNN-master/venv/lib/python3.8/site-packages/torchvision/transforms/transforms.py", line 179, in call return F.to_pil_image(pic, self.mode) File "/home/sharif/PyTorch-BayesianCNN-master/venv/lib/python3.8/site-packages/torchvision/transforms/functional.py", line 219, in to_pil_image raise ValueError('pic should be 2/3 dimensional. Got {} dimensions.'.format(pic.ndimension())) ValueError: pic should be 2/3 dimensional. Got 4 dimensions.