Benchmarking Suite for Loss Functions in T5 Training #36
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
…ss into loss-benchmarking
zausin33
reviewed
Jan 8, 2025
zausin33
approved these changes
Jan 8, 2025
…ss into loss-benchmarking
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Outdated suggestions cannot be applied.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.
This suite benchmarks different loss functions in neural network training across three scenarios:
1. Standalone loss computation
2. Model forward pass
3. Complete training step (including gradient updates)
Supported loss functions:
• Standard Cross Entropy (CE)
• CE with Number Token Loss (NTL) using MSE
• CE with NTL using Wasserstein distance
• CE with NTL using Absolute Difference
The suite includes utility functions for generating synthetic data, timing benchmarks, and logging results. It supports experimentation with different configurations and outputs CSV reports of benchmark statistics.