Zak acr #128
Merged
Zak acr #128
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Should now be ready with all the good changes |
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
Comment on lines
74
to
125
| # sparsifying filter-bank (SFB) module | ||
| class SFB(LIONmodel.LIONmodel): | ||
| def __init__(self, model_parameters): | ||
|
|
||
| super().__init__(model_parameters) | ||
| # FoE kernels | ||
| self.penalty = nn.Parameter((-12.0) * torch.ones(1)) | ||
| self.n_kernels = model_parameters.n_kernels | ||
| self.conv = nn.ModuleList( | ||
| [ | ||
| nn.Conv2d( | ||
| 1, | ||
| model_parameters.n_filters, | ||
| kernel_size=7, | ||
| stride=1, | ||
| padding=3, | ||
| bias=False, | ||
| ) | ||
| for i in range(self.n_kernels) | ||
| ] | ||
| ) | ||
| if model_parameters.L2net: | ||
| self.L2net = L2net() | ||
|
|
||
| @staticmethod | ||
| def default_parameters(): | ||
| param = LIONParameter() | ||
| param.n_kernels = 10 | ||
| param.n_filters = 32 | ||
| param.L2net = True | ||
| return param | ||
|
|
||
| def forward(self, x): | ||
| # compute the output of the FoE part | ||
| total_out = 0.0 | ||
| for kernel_idx in range(self.n_kernels): | ||
| x_out = torch.abs(self.conv[kernel_idx](x)) | ||
| x_out_flat = x_out.view(x.size(0), -1) | ||
| total_out += torch.sum(x_out_flat, dim=1) | ||
|
|
||
| total_out = total_out.view(x.size(0), -1) | ||
| out = (torch.nn.functional.softplus(self.penalty)) * total_out | ||
| if self.model_parameters.L2net: | ||
| out = out + self.L2net(x) | ||
| return out | ||
|
|
||
|
|
||
| class ACR(LIONmodel.LIONmodel): | ||
| def __init__(self, model_parameters: LIONParameter = None): | ||
|
|
||
| super().__init__(model_parameters) | ||
|
|
There was a problem hiding this comment.
I wrote this from Subhos code. I can't see it anywere (maybe Im blind) here. @Zakobian did you use this SBF in your code? why does your version of ACR not have it? or is it somewhere else that I am missing?
There was a problem hiding this comment.
SFB was in the ACR just for consistency check - not really part of the method
There was a problem hiding this comment.
@Zakobian so when you train it etc, you don't use it?
What do you mean "consistency check"?
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
AnderBiguri
reviewed
Aug 7, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
merge conflicts resolved