FIX: Fixed convergence issues by adding SlabGCN #22
+226
−56
Conversation
This file contains hidden or 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
Sign in
to join this conversation on GitHub.
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Outdated suggestions cannot be applied.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.
MultiGCNis difficult to train, probably because of the way it is designed. To make the network easier to train, a new architecture,SlabGCN, is added. The latter has a shared feed-forward neural network that takes in the concatenated embeddings of the partitioned GCNs as an input.SlabGCNin theModelclass in train.py. Also, removed "contributions" from everywhere for now.get_embeddingstoSlabGCNto get pooled vectors for each partition. This could be useful for model interpretation.With these changes the
SlabGCNmodel now gives an MAE of ~0.1-0.2 eV on the S calculation dataset.