Feed forward layer, frontend and encoder #53
Merged
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
added 3 commits
May 14, 2024 06:34
michelwi
reviewed
May 14, 2024
| x = x.transpose(1, 2) # torch 1d convolution is over last dim but we want time conv | ||
| x = self.conv(x).transpose(1, 2) | ||
|
|
||
| # these settings apparently apply stride correctly to the masking whatever the kernel size |
There was a problem hiding this comment.
Suggested change
| # these settings apparently apply stride correctly to the masking whatever the kernel size | |
| # change masking according to stride value |
do be confident about the implementation of others xP
There was a problem hiding this comment.
Actually I was not confident in my own implementation here. I had to choose kernel_size = 1 and padding = 0 to achieve correct masking, which is not how this function is supposed to be used I think.
There was a problem hiding this comment.
And with kernel_size = cfg.window_size and padding = get_same_padding(cfg.window_size) (i.e. the values given to Conv1d) it does not work?
There was a problem hiding this comment.
Weirdly enough yes. I came up with a failing test case in a new branch
i6_models/tests/test_masking.py
Line 7 in 55a6fd6
Co-authored-by: michelwi <[email protected]>
Atticus1806
reviewed
May 23, 2024
Co-authored-by: Benedikt Hilmes <[email protected]>
michelwi
approved these changes
May 23, 2024
Atticus1806
reviewed
May 24, 2024
Atticus1806
approved these changes
May 24, 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Outdated suggestions cannot be applied.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.
Implementation of a simple feed forward encoder that serves to generate good alignments in full-sum HMM training. Apart from the layers, it implements a simple convolutional front-end that acts as a feed-forward layer that takes a window of feature time frames into account.