Spec 10 15 #172
Merged
Spec 10 15 #172
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
dddg617
reviewed
Oct 31, 2023
gammagl/models/specformer.py
Outdated
| from tensorlayerx.model import WithLoss | ||
|
|
||
|
|
||
| def feature_normalize(x): |
There was a problem hiding this comment.
Did you use this function?
gammagl/models/specformer.py
Outdated
| rowsum = x.sum(axis=1, keepdims=True) | ||
| rowsum = np.clip(rowsum, 1, 1e10) | ||
| return x / rowsum | ||
| def normalize_graph(g): # input : g is adj matrix , return the L matrix of graph |
There was a problem hiding this comment.
I think these functions do not belong to this file, you may put these into the example/specformer/utils which means you need to create a new folder and put these functions into this folder.
gammagl/models/specformer.py
Outdated
|
|
||
| return train_index, valid_index, test_index | ||
| # utils | ||
| class DotProductAttention_tlx(tlx.nn.Module): |
There was a problem hiding this comment.
I think this class should be put into another file. You may put this into gammagl/models/attention.py as other users may use this class. And remember to add the docstring to explain the function and parameters.
gammagl/models/specformer.py
Outdated
| scores = tlx.ops.bmm(querys, tlx.nn.Transpose(perm=[0, 2, 1])(keys)) / math.sqrt(d) | ||
| self.attn_weights = tlx.nn.Softmax(axis=-1)(scores) | ||
| return tlx.ops.bmm(self.attn_weights, values) | ||
| class MultiHeadAttention_tlx(tlx.nn.Module): |
gammagl/models/specformer.py
Outdated
| eeig = tlx.cast(eeig, dtype=tlx.float32) | ||
| # print(eeig) | ||
| return self.eig_w(eeig) | ||
| class FeedForwardNetwork_tlx(tlx.nn.Module): |
There was a problem hiding this comment.
What is this doing? Why not directly use a two-layer linear layer in the following class?
dddg617
added a commit
that referenced
this pull request
Jan 12, 2024
* add specformer.py and spec_trainer.py * add readme.md * integrate 3 directories into 1 python file. * [Model] Update models * nightly update * nightly update --------- Co-authored-by: dddg617 <[email protected]> Co-authored-by: dddg617 <[email protected]>
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.
add specformer.py ,spec_trainer.py and readme.md into GammaGL