microsoft / microsoft/TransformerCompression

Speed up test

Open
#179 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
460
Forks
58
Avg merge
8h 40m
Merged PRs (30d)
1

Description

Hi,

Wonderful work u've done.

Some Q about the speed up for the structure pruning.

Actually, I pruned a llama2-7b into 0.3 sparsity, and test the forward time. Strangely, it cost more time to complete the forward time.

And then I test the attention forward and the feed forward then, the time consuming is below

Image

first is the dense model, later is the 0.3 sparsity model, it takes more time to ffn,

don't u have this problem?

then I test on the next toy

Image

so strange about my server!

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

The issue mentions no files, tests, or entry points. Start by reproducing the dense-versus-pruned forward-time comparison and document the model, environment, pruning configuration, and benchmark method; done should identify whether the slowdown is expected or reveal a reproducible performance defect.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.