Fix bfloat16/float16/float32 options #1369
Merged
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
Summary:
There was some problems with previous implementation of bfloat16/float16/float32 since it does not
convert activation to the correct dtype after quantization, this PR fixes it
Test Plan:
llama:
```
python generate.py --checkpoint_path $CHECKPOINT_PATH/$MODEL_REPO/model.pth --compile --compile_prefill --quantization autoquant-fp
```
same2:
```
server:
python server.py ~/checkpoints/sam2 large --port 8000 --host localhost --fast --use_autoquant
client:
time xargs -I {} curl -s -w "\n" -X POST http://localhost:8000/upload_rle -F 'image=@{}' < sav_val_image_paths_shuf_1000 > results/sav_val_masks_baseline_shuf_1000
```
Reviewers:
Subscribers:
Tasks:
Tags:
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1369
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 7463228 with merge base ed76e9c ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
cpuhrsch
approved these changes
Dec 2, 2024
jerryzh168
added
topic: new feature
yanbing-j
pushed a commit
to yanbing-j/ao
that referenced
this pull request
Dec 9, 2024
…ytorch#1369) * Only set up during the first sample * Cleaner
amdfaa
pushed a commit
that referenced
this pull request
Jan 10, 2025
* Fix bfloat16/float16/float32 options
Summary:
There was some problems with previous implementation of bfloat16/float16/float32 since it does not
convert activation to the correct dtype after quantization, this PR fixes it
Test Plan:
llama:
```
python generate.py --checkpoint_path $CHECKPOINT_PATH/$MODEL_REPO/model.pth --compile --compile_prefill --quantization autoquant-fp
```
same2:
```
server:
python server.py ~/checkpoints/sam2 large --port 8000 --host localhost --fast --use_autoquant
client:
time xargs -I {} curl -s -w "\n" -X POST http://localhost:8000/upload_rle -F 'image=@{}' < sav_val_image_paths_shuf_1000 > results/sav_val_masks_baseline_shuf_1000
```
Reviewers:
Subscribers:
Tasks:
Tags:
* ruff
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.
Summary:
There was some problems with previous implementation of bfloat16/float16/float32 since it does not convert activation to the correct dtype after quantization, this PR fixes it
Test Plan:
llama:
same2:
Reviewers:
Subscribers:
Tasks:
Tags: