Hi, This is really great work. I want to try your work but facing this error while running the evaluation.
Steps I did:
Downloaded and renamed the Model as GuideDepth bcoz that is what the model_builder function accepts.
Downloaded the Kiiti dataset(which you have provided) extracted it and created a tests folder in it and moved all the NumPy files to the tests folder.
I was getting an error of ' AttributeError: 'AverageMeter' object has no attribute 'sum_rmse_log' ' so initialized self.sum_rmse_log=0 in metrics.py
Run the code with this arguments python3 main.py --eval --dataset kitti --resolution full --model GuideDepth --test_path KITTI_test/ Now I am getting following error. Can you please help me with this ?
Namespace(batch_size=8, data_path='kitti_comb', dataset='kitti', eval_mode='alhashim', evaluate=True, learning_rate=0.0001, load_checkpoint='', model='GuideDepth', num_epochs=20, num_workers=2, resolution='full', save_checkpoint='./checkpoints', save_results='./results', scheduler_step_size=15, test_path='KITTI_test/', train=False, weights_path=None) Resolution for Eval: (384, 1280) Maximum Depth of Dataset: 80.0 Traceback (most recent call last): File "main.py", line 113, in <module> main() File "main.py", line 110, in main evaluation_module.evaluate() File "/home/control/GuidedDecoding/evaluate.py", line 88, in evaluate data = self.to_tensor(packed_data) File "/home/control/GuidedDecoding/data/transforms.py", line 89, in __call__ image, depth = transformation(image), transformation(depth) File "/home/control/anaconda3/envs/torch_0.4.1/lib/python3.7/site-packages/torchvision/transforms/transforms.py", line 76, in __call__ return F.to_tensor(pic) File "/home/control/anaconda3/envs/torch_0.4.1/lib/python3.7/site-packages/torchvision/transforms/functional.py", line 48, in to_tensor img = torch.from_numpy(pic.transpose((2, 0, 1))) ValueError: axes don't match array
Hi, This is really great work. I want to try your work but facing this error while running the evaluation. Steps I did:
python3 main.py --eval --dataset kitti --resolution full --model GuideDepth --test_path KITTI_test/
Now I am getting following error. Can you please help me with this ?Namespace(batch_size=8, data_path='kitti_comb', dataset='kitti', eval_mode='alhashim', evaluate=True, learning_rate=0.0001, load_checkpoint='', model='GuideDepth', num_epochs=20, num_workers=2, resolution='full', save_checkpoint='./checkpoints', save_results='./results', scheduler_step_size=15, test_path='KITTI_test/', train=False, weights_path=None) Resolution for Eval: (384, 1280) Maximum Depth of Dataset: 80.0 Traceback (most recent call last): File "main.py", line 113, in <module> main() File "main.py", line 110, in main evaluation_module.evaluate() File "/home/control/GuidedDecoding/evaluate.py", line 88, in evaluate data = self.to_tensor(packed_data) File "/home/control/GuidedDecoding/data/transforms.py", line 89, in __call__ image, depth = transformation(image), transformation(depth) File "/home/control/anaconda3/envs/torch_0.4.1/lib/python3.7/site-packages/torchvision/transforms/transforms.py", line 76, in __call__ return F.to_tensor(pic) File "/home/control/anaconda3/envs/torch_0.4.1/lib/python3.7/site-packages/torchvision/transforms/functional.py", line 48, in to_tensor img = torch.from_numpy(pic.transpose((2, 0, 1))) ValueError: axes don't match array
Thank you.