Attempt to change all Numpy calls to Torch calls #357
Conversation
… and not numpy 🧑🔬
| @@ -36,7 +36,7 @@ def get_cam_weights(self, | |||
| target_layers: List[torch.nn.Module], | |||
| targets: List[torch.nn.Module], | |||
| activations: torch.Tensor, | |||
| grads: torch.Tensor) -> np.ndarray: | |||
| grads: torch.Tensor) -> torch.Tensor: | |||
There was a problem hiding this comment.
For now Im trying to do minimal changes to create a proof of concept I can run.
Ive left this as a draft PR and its still definitely a WIP
I just find the PR user interface is great to observe changes
| img = img - np.min(img) | ||
| img = img / (1e-7 + np.max(img)) | ||
| img = img - torch.min(img) | ||
| img = img / (1e-7 + torch.max(img)) |
There was a problem hiding this comment.
The cv2 resize will need work to be done via a torch tensor. Will investigate once I get the concept working
pytorch_grad_cam/utils/image.py
Outdated
| result.append(img) | ||
| result = np.float32(result) | ||
| result = torch.tensor(np.array(result)).to(torch.float32) # TODO: Optimise this to use pre-initialised torch tensor |
There was a problem hiding this comment.
This is quite crude. My aim is to get torch working first. Then to go back to these hacks and get them optimal with a torch tensor approach once I know the rest of the changes to torch tensors are working.
Also helps identify the problem areas that need more attention.
None of this work is release ready
|
Sorry about the deluge of commits. If its an issue please let me know. I have remote access to the system Im running the library on and tend to commit in-order to execute my code there. Its part of my workflow but I can change it - if I generate too much noise |
| img = cv2.resize(img, target_size) | ||
| result.append(img) | ||
| result = np.float32(result) | ||
| # Disabled the target_size scaling for now |
There was a problem hiding this comment.
cv2.resize is a strange function. Looking online for a torch alternative brings up a few results. All have pros and cons are not exactly the same (including the dimension transpose).
This method is used in a few places in the code base. However for the GradCam class which Im using for timing tests its primarily used for the transpose (AFAIK), which isnt necessary here (or at least Im getting sane results my side).
Will need to be fixed if this change to a pure torch approach is used
| @@ -8,7 +8,7 @@ | |||
|
|
|||
| setuptools.setup( | |||
| name='grad-cam', | |||
| version='1.4.6', | |||
| version='1.4.7', | |||
There was a problem hiding this comment.
I just did this so on my compute node I know which version Im working with. In terms of what the real version bump would be Im open to a more major potential bump - as the final version of these changes would be substantial and potentially breaking
| cam = cam[:, 0, :, :] | ||
| cams.append(cam) | ||
| cams.append(cam) # TODO: Handle this for torch tensors |
There was a problem hiding this comment.
Basically pre-initialise a tensor. Ive found that to be drastically faster that lists when dealing with cuda / non-cpu devices
|
For now Im just modifying the code as I need to get the last of the results I need for my masters dissertation ... Ill have to cleanup later on |
What does this PR do?
Extremely experimental and proof of concept. Many areas may need improvement like resizing images