Commit Graph

56 Commits

Author SHA1 Message Date
OlivierDehaene 35509ff5de
chore: update to torch 2.1.0 (#1182)
Close #1142
2023-11-23 13:38:50 +01:00
Nicolas Patry 5ba53d44a1
Fixing eetq dockerfile. (#1081)
# What does this PR do?

Fixes #1079 
<!--
Congratulations! You've made it this far! You're not quite done yet
though.

Once merged, your PR is going to appear in the release notes with the
title you set, so make sure it's a great title that fully reflects the
extent of your awesome contribution.

Then, please replace this with a description of the change and which
issue is fixed (if applicable). Please also include relevant motivation
and context. List any dependencies (if any) that are required for this
change.

Once you're done, someone will review your PR shortly (see the section
"Who can review?" below to tag some potential reviewers). They may
suggest changes to make the code even better. If no one reviewed your PR
after a week has passed, don't hesitate to post a new comment
@-mentioning the same persons---sometimes notifications get lost.
-->

<!-- Remove if not applicable -->

Fixes # (issue)


## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Did you read the [contributor
guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests),
      Pull Request section?
- [ ] Was this discussed/approved via a Github issue or the
[forum](https://discuss.huggingface.co/)? Please add a link
      to it if that's the case.
- [ ] Did you make sure to update the documentation with your changes?
Here are the
[documentation
guidelines](https://github.com/huggingface/transformers/tree/main/docs),
and
[here are tips on formatting
docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation).
- [ ] Did you write any new necessary tests?


## Who can review?

Anyone in the community is free to review the PR once the tests have
passed. Feel free to tag
members/contributors who may be interested in your PR.

<!-- Your PR will be replied to more quickly if you can figure out the
right person to tag with @


@OlivierDehaene OR @Narsil

 -->
2023-09-29 11:19:06 +02:00
oOraph ae623b8d2d
Install curl to be able to perform more advanced healthchecks (#1033)
# What does this PR do?

Install curl within base image, negligible regarding the image volume
and will allow to easily perform a better health check. Not sure about
the failing github actions though. Should I fix something ?

Signed-off-by: Raphael <oOraph@users.noreply.github.com>
Co-authored-by: Raphael <oOraph@users.noreply.github.com>
2023-09-26 15:23:47 +02:00
Nicolas Patry c5de7cd886
Add AWQ quantization inference support (#1019) (#1054)
# Add AWQ quantization inference support

Fixes
https://github.com/huggingface/text-generation-inference/issues/781

This PR (partially) adds support for AWQ quantization for inference.
More information on AWQ [here](https://arxiv.org/abs/2306.00978). In
general, AWQ is faster and more accurate than GPTQ, which is currently
supported by TGI.

This PR installs 4-bit GEMM custom CUDA kernels released by AWQ authors
(in `requirements.txt`, just one line change).

Quick way to test this PR would be bring up TGI as follows:

```
text-generation-server download-weights abhinavkulkarni/codellama-CodeLlama-7b-Python-hf-w4-g128-awq

text-generation-launcher \
--huggingface-hub-cache ~/.cache/huggingface/hub/ \
--model-id abhinavkulkarni/codellama-CodeLlama-7b-Python-hf-w4-g128-awq \
--trust-remote-code --port 8080 \
--max-input-length 2048 --max-total-tokens 4096 --max-batch-prefill-tokens 4096 \
--quantize awq
```

Please note:
* This PR was tested with FlashAttention v2 and vLLM.
* This PR adds support for AWQ inference, not quantizing the models.
That needs to be done outside of TGI, instructions

[here](f084f40bd9).
* This PR only adds support for `FlashLlama` models for now.
* Multi-GPU setup has not been tested. 
* No integration tests have been added so far, will add later if
maintainers are interested in this change.
* This PR can be tested on any of the models released

[here](https://huggingface.co/abhinavkulkarni?sort_models=downloads#models).

Please refer to the linked issue for benchmarks for

[abhinavkulkarni/meta-llama-Llama-2-7b-chat-hf-w4-g128-awq](https://huggingface.co/abhinavkulkarni/meta-llama-Llama-2-7b-chat-hf-w4-g128-awq)
vs

[TheBloke/Llama-2-7b-Chat-GPTQ](https://huggingface.co/TheBloke/Llama-2-7b-Chat-GPTQ).

Please note, AWQ has released faster (and in case of Llama, fused)
kernels for 4-bit GEMM, currently at the top of the `main` branch at
https://github.com/mit-han-lab/llm-awq, but this PR uses an older commit
that has been tested to work. We can switch to latest commit later on.

## Who can review?

@OlivierDehaene OR @Narsil

---------



# What does this PR do?

<!--
Congratulations! You've made it this far! You're not quite done yet
though.

Once merged, your PR is going to appear in the release notes with the
title you set, so make sure it's a great title that fully reflects the
extent of your awesome contribution.

Then, please replace this with a description of the change and which
issue is fixed (if applicable). Please also include relevant motivation
and context. List any dependencies (if any) that are required for this
change.

Once you're done, someone will review your PR shortly (see the section
"Who can review?" below to tag some potential reviewers). They may
suggest changes to make the code even better. If no one reviewed your PR
after a week has passed, don't hesitate to post a new comment
@-mentioning the same persons---sometimes notifications get lost.
-->

<!-- Remove if not applicable -->

Fixes # (issue)


## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Did you read the [contributor
guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests),
      Pull Request section?
- [ ] Was this discussed/approved via a Github issue or the
[forum](https://discuss.huggingface.co/)? Please add a link
      to it if that's the case.
- [ ] Did you make sure to update the documentation with your changes?
Here are the
[documentation
guidelines](https://github.com/huggingface/transformers/tree/main/docs),
and
[here are tips on formatting
docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation).
- [ ] Did you write any new necessary tests?


## Who can review?

Anyone in the community is free to review the PR once the tests have
passed. Feel free to tag
members/contributors who may be interested in your PR.

<!-- Your PR will be replied to more quickly if you can figure out the
right person to tag with @


@OlivierDehaene OR @Narsil

 -->

---------

Co-authored-by: Abhinav M Kulkarni <abhinavkulkarni@gmail.com>
Co-authored-by: Abhinav Kulkarni <abhinav@concentric.ai>
2023-09-25 15:31:27 +02:00
Nicolas Patry 6ec5288ab7
This should prevent the PyTorch overriding. (#767)
# What does this PR do?

<!--
Congratulations! You've made it this far! You're not quite done yet
though.

Once merged, your PR is going to appear in the release notes with the
title you set, so make sure it's a great title that fully reflects the
extent of your awesome contribution.

Then, please replace this with a description of the change and which
issue is fixed (if applicable). Please also include relevant motivation
and context. List any dependencies (if any) that are required for this
change.

Once you're done, someone will review your PR shortly (see the section
"Who can review?" below to tag some potential reviewers). They may
suggest changes to make the code even better. If no one reviewed your PR
after a week has passed, don't hesitate to post a new comment
@-mentioning the same persons---sometimes notifications get lost.
-->

<!-- Remove if not applicable -->

Fixes # (issue)


## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Did you read the [contributor
guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests),
      Pull Request section?
- [ ] Was this discussed/approved via a Github issue or the
[forum](https://discuss.huggingface.co/)? Please add a link
      to it if that's the case.
- [ ] Did you make sure to update the documentation with your changes?
Here are the
[documentation
guidelines](https://github.com/huggingface/transformers/tree/main/docs),
and
[here are tips on formatting
docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation).
- [ ] Did you write any new necessary tests?


## Who can review?

Anyone in the community is free to review the PR once the tests have
passed. Feel free to tag
members/contributors who may be interested in your PR.

<!-- Your PR will be replied to more quickly if you can figure out the
right person to tag with @


@OlivierDehaene OR @Narsil

 -->
2023-08-03 21:54:39 +02:00
Nicolas Patry 92bb56b0c1
Local gptq support. (#738)
# What does this PR do?

Redoes #719

<!--
Congratulations! You've made it this far! You're not quite done yet
though.

Once merged, your PR is going to appear in the release notes with the
title you set, so make sure it's a great title that fully reflects the
extent of your awesome contribution.

Then, please replace this with a description of the change and which
issue is fixed (if applicable). Please also include relevant motivation
and context. List any dependencies (if any) that are required for this
change.

Once you're done, someone will review your PR shortly (see the section
"Who can review?" below to tag some potential reviewers). They may
suggest changes to make the code even better. If no one reviewed your PR
after a week has passed, don't hesitate to post a new comment
@-mentioning the same persons---sometimes notifications get lost.
-->

<!-- Remove if not applicable -->

Fixes # (issue)


## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Did you read the [contributor
guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests),
      Pull Request section?
- [ ] Was this discussed/approved via a Github issue or the
[forum](https://discuss.huggingface.co/)? Please add a link
      to it if that's the case.
- [ ] Did you make sure to update the documentation with your changes?
Here are the
[documentation
guidelines](https://github.com/huggingface/transformers/tree/main/docs),
and
[here are tips on formatting
docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation).
- [ ] Did you write any new necessary tests?


## Who can review?

Anyone in the community is free to review the PR once the tests have
passed. Feel free to tag
members/contributors who may be interested in your PR.

<!-- Your PR will be replied to more quickly if you can figure out the
right person to tag with @


@OlivierDehaene OR @Narsil

 -->
2023-07-31 10:32:52 +02:00
OlivierDehaene 2efd46ef95 fix(server): fix missing datasets in quantize 2023-07-27 14:50:45 +02:00
Nicolas Patry d5b5bc750f
feat(server): Add exllama GPTQ CUDA kernel support #553 (#666)
Just trying to get the integration tests to pass.


# What does this PR do?

<!--
Congratulations! You've made it this far! You're not quite done yet
though.

Once merged, your PR is going to appear in the release notes with the
title you set, so make sure it's a great title that fully reflects the
extent of your awesome contribution.

Then, please replace this with a description of the change and which
issue is fixed (if applicable). Please also include relevant motivation
and context. List any dependencies (if any) that are required for this
change.

Once you're done, someone will review your PR shortly (see the section
"Who can review?" below to tag some potential reviewers). They may
suggest changes to make the code even better. If no one reviewed your PR
after a week has passed, don't hesitate to post a new comment
@-mentioning the same persons---sometimes notifications get lost.
-->

<!-- Remove if not applicable -->

Fixes # (issue)


## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Did you read the [contributor
guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests),
      Pull Request section?
- [ ] Was this discussed/approved via a Github issue or the
[forum](https://discuss.huggingface.co/)? Please add a link
      to it if that's the case.
- [ ] Did you make sure to update the documentation with your changes?
Here are the
[documentation
guidelines](https://github.com/huggingface/transformers/tree/main/docs),
and
[here are tips on formatting
docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation).
- [ ] Did you write any new necessary tests?


## Who can review?

Anyone in the community is free to review the PR once the tests have
passed. Feel free to tag
members/contributors who may be interested in your PR.

<!-- Your PR will be replied to more quickly if you can figure out the
right person to tag with @


@OlivierDehaene OR @Narsil

 -->

---------

Co-authored-by: Felix Marty <9808326+fxmarty@users.noreply.github.com>
2023-07-21 10:59:00 +02:00
OlivierDehaene 3b71c38558
feat(server): flash attention v2 (#624) 2023-07-18 16:21:18 +02:00
OlivierDehaene e28a809004
v0.9.0 (#525) 2023-07-01 19:25:41 +02:00
OlivierDehaene e74bd41e0f
feat(server): add paged attention to flash models (#516)
Closes #478
2023-06-30 19:09:59 +02:00
Nicolas Patry aefde28b45
feat(server): Add inference support for GPTQ (llama + falcon tested) + Quantization script (#438)
Let's start discussing implementation.

- Need to expose the quantization scripts (either included here or add
doc on how to use https://github.com/qwopqwop200/GPTQ-for-LLaMa)
- Make sure GPTQ works for multiple models (priority to Falcon).

Currently it means that every place we use `get_{tensor|sharded}` to
check for quantization.

My idea is to reintegrate as much as possible into `utils/layer.py` by
expanding `load_multi` to be a bit more generic.
This might require some thinking, but ultimately the
`qweight,qzeros,scales,g_idx` should be in a single place, and
independant of bias presence.

# What does this PR do?

<!--
Congratulations! You've made it this far! You're not quite done yet
though.

Once merged, your PR is going to appear in the release notes with the
title you set, so make sure it's a great title that fully reflects the
extent of your awesome contribution.

Then, please replace this with a description of the change and which
issue is fixed (if applicable). Please also include relevant motivation
and context. List any dependencies (if any) that are required for this
change.

Once you're done, someone will review your PR shortly (see the section
"Who can review?" below to tag some potential reviewers). They may
suggest changes to make the code even better. If no one reviewed your PR
after a week has passed, don't hesitate to post a new comment
@-mentioning the same persons---sometimes notifications get lost.
-->

<!-- Remove if not applicable -->

Fixes # (issue)


## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Did you read the [contributor
guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests),
      Pull Request section?
- [ ] Was this discussed/approved via a Github issue or the
[forum](https://discuss.huggingface.co/)? Please add a link
      to it if that's the case.
- [ ] Did you make sure to update the documentation with your changes?
Here are the
[documentation
guidelines](https://github.com/huggingface/transformers/tree/main/docs),
and
[here are tips on formatting
docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation).
- [ ] Did you write any new necessary tests?


## Who can review?

Anyone in the community is free to review the PR once the tests have
passed. Feel free to tag
members/contributors who may be interested in your PR.

<!-- Your PR will be replied to more quickly if you can figure out the
right person to tag with @


@OlivierDehaene OR @Narsil

 -->

---------

Co-authored-by: Ubuntu <ubuntu@ip-172-31-41-161.ec2.internal>
Co-authored-by: OlivierDehaene <olivier@huggingface.co>
2023-06-26 12:27:01 +02:00
Nicolas Patry abd58ff82c
feat(server): Rework model loading (#344)
# What does this PR do?

Reworked the loading logic. Idea is to use cleaner loading code:

- Remove need for `no_init_weights`
- Remove all weird `bnb_linear` and `load_weights` and
`post_load_weights`.

New code layout:

- New class `Weights` in charge of handling loading the weights from
multiple files into appropiate tensors (potentially sharded)
- TP layers now are "shells", they contain the code to know what kind of
sharding we need + eventual `all_reduce`. They do not inherit from
linear, but they contain some kind of Linear instead
- the contained linear can be either FastLinear, BnbLinear or GPTq
Linear next.
- All modeling code is explictly made for sharding, process group is
just no-ops for non sharded code (removes a lot of test cases)

![Screenshot from 2023-05-19
23-19-59](https://github.com/huggingface/text-generation-inference/assets/204321/9a802654-74a3-488c-87a8-073743a6143f)

---------

Co-authored-by: Ubuntu <ubuntu@ip-172-31-41-161.taildb5d.ts.net>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-41-161.ec2.internal>
Co-authored-by: OlivierDehaene <olivier@huggingface.co>
Co-authored-by: OlivierDehaene <23298448+OlivierDehaene@users.noreply.github.com>
2023-06-08 14:51:52 +02:00
OlivierDehaene 22c4fd07ab fix(docker): use ubuntu20.04 2023-05-12 18:38:59 +02:00
OlivierDehaene 119f7e0687 fix(docker): remove quantize default 2023-05-12 17:56:32 +02:00
OlivierDehaene 8a8f43410d
chore(docker): use nvidia base image (#318) 2023-05-12 17:32:40 +02:00
OlivierDehaene 35ab6cfcf1 fix(docker): remove CUDA_VERSION 2023-05-10 16:16:06 +02:00
OlivierDehaene 1585404464
fix(docker): remove nvidia require cuda env (#310) 2023-05-10 15:29:21 +02:00
OlivierDehaene 49cffad1bc
fix(docker): fix nvidia env vars (#305) 2023-05-09 19:02:52 +02:00
OlivierDehaene bc5c07231e
fix(docker): fix docker build (#299) 2023-05-09 14:39:59 +02:00
OlivierDehaene e250282213
feat(docker): add benchmarking tool to docker image (#298) 2023-05-09 13:19:31 +02:00
OlivierDehaene e9b01b3433
fix(dockerfile): fix nvidia env vars (#297)
Fixes #291
2023-05-09 12:36:02 +02:00
Nicolas Patry 411b0d4e1f
chore(github): add templates (#264) 2023-05-02 15:43:19 +02:00
OlivierDehaene 593a563414
feat(docker): add nvidia env vars (#255) 2023-04-27 19:18:33 +02:00
OlivierDehaene 98a3e0d135
chore(server): update huggingface-hub (#227) 2023-04-24 15:57:13 +02:00
OlivierDehaene 97df0c7bc0
misc: update to rust 1.69 (#221) 2023-04-21 21:00:30 +02:00
OlivierDehaene b6ee0ec7b0
feat(router): add git sha to info route (#208) 2023-04-19 21:36:59 +02:00
OlivierDehaene 6837b2eb77
fix(docker): remove unused dependencies (#205) 2023-04-19 19:39:31 +02:00
OlivierDehaene 5d27f5259b
fix(server): fix hf_transfer issue with private repos (#203) 2023-04-19 17:36:16 +02:00
OlivierDehaene 7a1ba58557
fix(docker): fix docker image dependencies (#187) 2023-04-17 00:26:47 +02:00
OlivierDehaene 379c5c4da2
fix(docker): revert dockerfile changes (#186) 2023-04-14 19:30:30 +02:00
OlivierDehaene f9047562d0
fix(docker): fix image (#185) 2023-04-14 18:58:38 +02:00
OlivierDehaene 1bb394631d
fix(docker): fix docker image (#184) 2023-04-14 17:31:13 +02:00
OlivierDehaene 53ee09c0b0
fea(dockerfile): better layer caching (#159) 2023-04-14 10:12:21 +02:00
OlivierDehaene 1883d8ecde
feat(docker): improve flash_attention caching (#160) 2023-04-09 19:59:16 +02:00
OlivierDehaene d503e8f09d
feat: aws sagemaker compatible image (#147)
The only difference is that now it pushes to
registry.internal.huggingface.tech/api-inference/community/text-generation-inference/sagemaker:...
instead of
registry.internal.huggingface.tech/api-inference/community/text-generation-inference:sagemaker-...

---------

Co-authored-by: Philipp Schmid <32632186+philschmid@users.noreply.github.com>
2023-03-29 21:38:30 +02:00
OlivierDehaene 05e9a796cc
feat(server): flash neoX (#133) 2023-03-24 14:02:14 +01:00
OlivierDehaene e3ded361b2
feat(ci): improve CI speed (#94) 2023-03-03 15:07:27 +01:00
OlivierDehaene 17bc841b1b
feat(server): enable hf-transfer (#76) 2023-02-18 14:04:11 +01:00
OlivierDehaene 9af454142a
feat: add distributed tracing (#62) 2023-02-13 13:02:45 +01:00
OlivierDehaene 1ad3250b89
fix(docker): increase shm size (#60) 2023-02-08 17:53:33 +01:00
OlivierDehaene 20c3c5940c
feat(router): refactor API and add openAPI schemas (#53) 2023-02-03 12:43:37 +01:00
OlivierDehaene 13e7044ab7
fix(dockerfile): fix docker build (#32) 2023-01-24 19:52:39 +01:00
OlivierDehaene ab2ad91da3
fix(docker): fix api-inference deployment (#30) 2023-01-23 17:33:08 +01:00
OlivierDehaene f9d0ec376a
feat(docker): Make the image compatible with api-inference (#29) 2023-01-23 17:11:27 +01:00
OlivierDehaene 6c781025ae feat(rust): Update to 1.65 2022-11-14 13:59:56 +01:00
OlivierDehaene fa43fb71be fix(server): Fix Transformers fork version 2022-11-08 17:42:38 +01:00
OlivierDehaene 4236e41b0d feat(server): Improved doc 2022-11-07 12:53:56 +01:00
OlivierDehaene b3b7ea0d74 feat: Use json formatter by default in docker image 2022-11-02 17:29:56 +01:00
OlivierDehaene 3cf6368c77 feat(server): Support all AutoModelForCausalLM on a best effort basis 2022-10-28 19:24:00 +02:00