[Inference] Add a8w8(fp8) a8w8c8(int8) quant_type support #9032
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
2. add llama3.1 and qwen2 ptq config 3. update quantization.md
|
Thanks for your contribution! |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #9032 +/- ##
===========================================
+ Coverage 53.81% 54.01% +0.19%
===========================================
Files 652 652
Lines 104356 105208 +852
===========================================
+ Hits 56155 56823 +668
- Misses 48201 48385 +184 ☔ View full report in Codecov by Sentry. |
DrownFish19
changed the title
yuanlehome
reviewed
Aug 28, 2024
| @@ -220,6 +228,12 @@ python run_finetune.py ./config/llama/ptq_argument.json | |||
|
|
|||
| # GPTQ 量化启动命令参考 | |||
| python run_finetune.py ./config/llama/ptq_argument.json | |||
|
|
|||
| # W8A8C8(INT)量化启动命令参考 | |||
| python run_finetune.py ./config/llama/ptq_c8_argument.json | |||
| @@ -0,0 +1,138 @@ | |||
| # Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved. | |||
Comment on lines
+47
to
+49
| # print( | ||
| # f"{index/len(subject_list)} Inference starts at {run_date} on {args.model_name_or_path} with subject of {subject_name}!" | ||
| # ) |
| @@ -0,0 +1,61 @@ | |||
| # Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved. | |||
| @@ -0,0 +1,191 @@ | |||
| # Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved. | |||
| @@ -0,0 +1,94 @@ | |||
| # Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved. | |||
| import numpy as np | ||
| import paddle | ||
|
|
||
| # from paddleslim.quant.observers.channel_wise import ChannelWiseObserver |
| @@ -0,0 +1,105 @@ | |||
| # Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved. | |||
| @@ -0,0 +1,55 @@ | |||
| # Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved. | |||
| # from paddle.quantization.factory import ObserverFactory | ||
| from experimental.layers.cache_kv import CacheKVMatMul | ||
|
|
||
| # from paddleslim.quant.observers.mse import MSEObserverLayer |
Mangodadada
pushed a commit
to Mangodadada/PaddleNLP
that referenced
this pull request
Sep 10, 2024
…le#9032) * 1. add a8w8(fp8) a8w8c8(int8) quant_type support 2. add llama3.1 and qwen2 ptq config 3. update quantization.md * fix load_quant_model bug * fix load quant bug * update ll/README.md
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.
PR types
New features
PR changes
APIs | Docs
Description