DLRM training for MLPerf v1.0 submission. not merged. #163
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
2)LAMB to enable large batch size.
EikanWang
pushed a commit
that referenced
this pull request
Oct 4, 2021
Peach-He
reviewed
Jan 13, 2022
Comment on lines
+830
to
+839
| + optimizer_dense = optimizers[0][0]([ | ||
| + {"params": [p for emb in dlrm.emb_dense for p in emb.parameters()], "lr": args.learning_rate}, | ||
| + {"params": dlrm.bot_l.parameters(), "lr": args.learning_rate}, | ||
| + {"params": dlrm.top_l.parameters(), "lr": args.learning_rate} | ||
| + ], lr=args.lamblr, bf16=args.bf16) | ||
| + optimizer_sparse = optimizers[1]([ | ||
| + {"params": [p for emb in dlrm.emb_sparse for p in emb.parameters()], | ||
| + "lr": args.learning_rate / ext_dist.my_size}, | ||
| + ], lr=args.learning_rate) | ||
| + optimizer = (optimizer_dense, optimizer_sparse) |
There was a problem hiding this comment.
I want to upgrade this DLRM optimization to latest IPEX 1.10, any example on optimizing multi optimizers with ipex?
There was a problem hiding this comment.
There is no such case now.
# for free
to join this conversation on GitHub.
Already have an account?
# 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Outdated suggestions cannot be applied.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.
1)Vertical split embedding to scale-out to much more ranks.
2)LAMB to enable large batch size.