Commit Graph

621 Commits

Author SHA1 Message Date
OlivierDehaene ff42d33e99
Revert license to Apache 2.0 (#1714)
Reverts huggingface/text-generation-inference#725

---------

Co-authored-by: Julien Chaumond <julien@huggingface.co>
2024-04-08 15:06:16 +02:00
oOraph 53c2c3dbc7
Regenerate ld.so.cache (#1708)
fixes
https://github.com/huggingface/text-generation-inference/issues/1711

Signed-off-by: Raphael Glon <oOraph@users.noreply.github.com>
Co-authored-by: Raphael Glon <oOraph@users.noreply.github.com>
2024-04-08 08:52:10 +02:00
Nicolas Patry 8dca3b04f8
Force weights_only (before fully breaking pickle files anyway). (#1710)
# 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

 -->
2024-04-05 19:23:57 +02:00
Nicolas Patry f9958ee191
Fixing cohere tokenizer. (#1697) 2024-04-05 16:44:19 +02:00
Nicolas Patry 5062fda4ff
Push users to streaming in the readme. (#1698) 2024-04-05 16:44:10 +02:00
Nicolas Patry c7e570e59d
Pickle conversion now requires `--trust-remote-code`. (#1704)
# 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

 -->
2024-04-05 13:32:53 +02:00
Nicolas Patry 99874eae74
Add cuda graphs sizes and make it default. (#1703)
# What does this PR do?

```
text-generation-launcher --model-id XXX # Uses cuda graphs by default
text-generation-launcher --model-id XXX --cuda-graphs "1,2"  #Restrict the number of cuda graphs which saves VRAM
text-generation-launcher --model-id XXX --cuda-graphs "0"  # Disabling it entirely
```
<!--
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

 -->
2024-04-04 23:01:56 +02:00
OlivierDehaene 4ee0a0c401
v1.4.5 (#1686) 2024-03-29 19:17:24 +01:00
OlivierDehaene f04255c694
feat: Add dbrx support (#1685)
Close #1679
2024-03-29 18:49:36 +01:00
drbh 762dbf3f19
fix: handle batches with and without grammars (#1676)
This PR correctly handles batches with a mixture of constrained and non
constrained generations.

Currently if batch contains mixed generations the generation will throw
an error because it will incorrectly attempt to constrain a request with
an empty grammar.

We now handled `None` grammars and only apply the mask if needed

Fixes:
https://github.com/huggingface/text-generation-inference/issues/1643
2024-03-28 12:02:01 -04:00
drbh 818aee37e5
fix: adjust logprob response logic (#1682)
This PR fixes a bug with `ChatCompletionLogprobs` where if
`top_tokens.len() == 0` empty results were returned.

```bash
 curl http://localhost:3000/v1/chat/completions \
    -X POST \
    -H 'Content-Type: application/json' \
    -d '{
  "model": "tgi",
  "logprobs": true,
  "messages": [
    {
      "role": "user",
      "content": "What is deep learning?"
    }
  ],
  "stream": false,
  "max_tokens": 20
}'
```

response


```json
{"id":"","object":"text_completion","created":1711588522,"model":"google/gemma-2b-it","system_fingerprint":"1.4.4-native","choices":[{"index":0,"message":{"role":"assistant","content":"**Deep learning** is a subset of machine learning (ML) that emphasizes the creation of **artificial"},"logprobs":{"content":[{"token":"**","logprob":-0.22558594,"top_logprobs":[]},{"token":"Deep","logprob":-0.0014877319,"top_logprobs":[]},{"token":" learning","logprob":-0.12695312,"top_logprobs":[]},{"token":"**","logprob":-0.055664062,"top_logprobs":[]},{"token":" is","logprob":-0.00090026855,"top_logprobs":[]},{"token":" a","logprob":-0.006072998,"top_logprobs":[]},{"token":" subset","logprob":-2.25,"top_logprobs":[]},{"token":" of","logprob":-0.00031089783,"top_logprobs":[]},{"token":" machine","logprob":-0.091308594,"top_logprobs":[]},{"token":" learning","logprob":-0.00002348423,"top_logprobs":[]},{"token":" (","logprob":-1.671875,"top_logprobs":[]},{"token":"ML","logprob":-0.00040626526,"top_logprobs":[]},{"token":")","logprob":-0.00016212463,"top_logprobs":[]},{"token":" that","logprob":-0.13769531,"top_logprobs":[]},{"token":" emphasizes","logprob":-4.03125,"top_logprobs":[]},{"token":" the","logprob":-0.2890625,"top_logprobs":[]},{"token":" creation","logprob":-3.109375,"top_logprobs":[]},{"token":" of","logprob":-0.00024032593,"top_logprobs":[]},{"token":" **","logprob":-1.2265625,"top_logprobs":[]},{"token":"artificial","logprob":-0.10546875,"top_logprobs":[]}]},"finish_reason":"length"}],"usage":{"prompt_tokens":15,"completion_tokens":20,"total_tokens":35}}
```
2024-03-28 12:01:46 -04:00
OlivierDehaene 6c4496a1a3
v1.4.4 (#1668) 2024-03-22 18:44:05 +01:00
OlivierDehaene 1e9bcd9dd8
feat: cohere (#1660) 2024-03-22 17:59:25 +01:00
Nicolas Patry f171bdc823
Inline images for multimodal models. (#1666) 2024-03-22 17:14:54 +01:00
SeongBeomLEE 66914f7b19
fix: LlamaTokenizerFast to AutoTokenizer at flash_mistral.py (#1637)
# What does this PR do?

A few cases where you're using a mistral structure or mixtral structure
but not a llama tokenizer, why not make it to call the AutoTokenizer in
exception handling.

Similar PR #619

@Narsil
2024-03-22 17:13:13 +01:00
OlivierDehaene 08e9181418
feat: update client to 0.7 (#1667)
Close #1652
2024-03-22 17:10:56 +01:00
Nicolas Patry deb440b3a2
Repair idefics integration tests. (#1663)
# 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

 -->
2024-03-21 22:21:03 +01:00
Nicolas Patry ed29d6eeab
Remove unecessary cuda graph. (#1664)
# 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

 -->
2024-03-21 20:11:05 +01:00
drbh de6cb15fa5
fix: improve tool type, bump pydantic and outlines (#1650)
This PR resolves a couple 

- [X] adjusts the tool response to align with openai's tools response
type
- [X] bumps pydantic to `2.6.4` in all apps (resolves dependency issue
when running tests)
- [X] bump `outlines` version and fix import for new name
2024-03-21 12:45:56 -04:00
drbh 4f09c80cd8
fix: prefer spaces url over temp url (#1662)
This PR fixes the broken urls in the idefics tests causing CI to fail
2024-03-21 17:34:25 +01:00
drbh 6f15ac60b2
feat: support force downcast after FastRMSNorm multiply for Gemma (#1658)
This PR adds `force_downcast_after` to `FastRMSNorm.forward` which is
used in the Gemma model. References
https://github.com/huggingface/transformers/pull/29402 and
https://github.com/huggingface/transformers/pull/29729

Setting `force_downcast_after=True` will perform the `hidden_states *
weight` multiplication in f32 and then downcast to half. This differs
slightly from the current implementation which first casts the
`hidden_states` to a half and then multiples.
2024-03-21 10:25:11 +01:00
drbh dfbd9a39a2
feat: bump minijina and add test for core templates (#1626)
This PR bumps `minijinja` and adds tests for all core models as
identified by @xenova 🙏

Inspiration:
https://github.com/huggingface/huggingface.js/blob/main/packages/jinja/test/e2e.test.js

TODO:
- [X] add new test to iterate over known templates
- [X] add default templates
- [x] add custom templates
2024-03-20 09:13:46 -04:00
Sachin Varghese 0d72af5ab0
Fixing minor typo in documentation: supported hardware section (#1632)
# 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.
-->


## Before submitting
- [x] 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

 -->

Signed-off-by: Sachin Varghese <sachin.mathew31@gmail.com>
2024-03-18 14:33:58 +08:00
Lucain 23fba672e8
Fix index in ChatCompletionChunk (#1648)
Fix a small inconsistency compared the OpenAI's chat-completion behavior
(introduced in
https://github.com/huggingface/text-generation-inference/pull/1427 cc
@drbh). When using `stream=True`, each chunk has an `index` value in
`ChatCompletionChoice`. This index is not meant to be the index of the
generated token but the index of the choice, which is always 0 (since
TGI always return a single choice).

See https://platform.openai.com/docs/api-reference/chat/object:
> index _integer_
> The index of the choice in the list of choices.

---

So instead of 

```js
data:{"id":"","object":"text_completion","created":1710508199,"model":"HuggingFaceH4/zephyr-7b-beta","system_fingerprint":"1.4.3-sha-e6bb3ff","choices":[{"index":1,"delta":{"role":"assistant","content":"I"},"logprobs":null,"finish_reason":null}]}
data:{"id":"","object":"text_completion","created":1710508199,"model":"HuggingFaceH4/zephyr-7b-beta","system_fingerprint":"1.4.3-sha-e6bb3ff","choices":[{"index":2,"delta":{"role":"assistant","content":"'"},"logprobs":null,"finish_reason":null}]}
data:{"id":"","object":"text_completion","created":1710508199,"model":"HuggingFaceH4/zephyr-7b-beta","system_fingerprint":"1.4.3-sha-e6bb3ff","choices":[{"index":3,"delta":{"role":"assistant","content":"m"},"logprobs":null,"finish_reason":"length"}]}
```

if should return
```js
data:{"id":"","object":"text_completion","created":1710508199,"model":"HuggingFaceH4/zephyr-7b-beta","system_fingerprint":"1.4.3-sha-e6bb3ff","choices":[{"index":0,"delta":{"role":"assistant","content":"I"},"logprobs":null,"finish_reason":null}]}
data:{"id":"","object":"text_completion","created":1710508199,"model":"HuggingFaceH4/zephyr-7b-beta","system_fingerprint":"1.4.3-sha-e6bb3ff","choices":[{"index":0,"delta":{"role":"assistant","content":"'"},"logprobs":null,"finish_reason":null}]}
data:{"id":"","object":"text_completion","created":1710508199,"model":"HuggingFaceH4/zephyr-7b-beta","system_fingerprint":"1.4.3-sha-e6bb3ff","choices":[{"index":0,"delta":{"role":"assistant","content":"m"},"logprobs":null,"finish_reason":"length"}]}
```

**EDIT:** I also edited ToolCall.index to be always `0` (instead of the
generated token index) but for this one I'm actually unsure. It might be
the index of the tool in the array of tools? OpenAI's documentation
doesn't provide any information about it:
> index _integer_

---

I also noticed that in OpenAI's example, the last chunk doesn't have a
delta and is the only one that has a `finish_reason` returning. TGI is
slightly different since the last chunk has both the last delta (i.e.
the last generated token) + the finish reason. I don't think this is
worth fixing since it is not a requirement according to the docs/specs
(at least not that I know of).
2024-03-16 12:14:29 -04:00
abhishek thakur 0d9917f744
Update peft + transformers + accelerate + bnb + safetensors (#1646) 2024-03-15 13:23:26 +01:00
yuanwu2017 8a5bcba227
Upgrade nix version from 0.27.1 to 0.28.0 (#1638)
# What does this PR do?
Fix the following carsh when build the docker on Ubuntu22.04

```
error[E0432]: unresolved import `nix::sys::signal::Signal`
 --> launcher/src/main.rs:2:30
  |
2 | use nix::sys::signal::{self, Signal};
  |                              ^^^^^^ no `Signal` in `sys::signal`
  |
  = help: consider importing this type alias instead:
          ctrlc::Signal

error[E0432]: unresolved import `nix::unistd::Pid`
   --> launcher/src/main.rs:3:5
    |
3   | use nix::unistd::Pid;
    |     ^^^^^^^^^^^^^^^^ no `Pid` in `unistd`
    |
note: found an item that was configured out
   --> /usr/local/cargo/registry/src/index.crates.io-6f17d22bba15001f/nix-0.27.1/src/unistd.rs:183:12
    |
183 | pub struct Pid(pid_t);
    |            ^^^
    = note: the item is gated behind the `process` feature

error[E0425]: cannot find function `kill` in module `signal`

```



<!--
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

 -->

---------

Signed-off-by: yuanwu <yuan.wu@intel.com>
Co-authored-by: drbh <david.richard.holtz@gmail.com>
2024-03-12 08:37:33 -04:00
lewtun d3711a6646
Use a better model for the quick tour (#1639)
Falcon models are long superseded by better models like Zephyr and
OpenHermes. This PR updates the docs accordingly

# 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

 -->
2024-03-12 11:25:32 +01:00
drbh 7dbaf9e901
fix: correctly index into mask when applying grammar (#1618)
This PR fixes how the grammar mask is index when generating text and
adds a new test to ensure the grammars work with non flash models
2024-03-01 18:22:01 +01:00
drbh 7e08751378
fix: add missing stop parameter for chat request (#1619)
This PR adds the missing `stop` parameter to the `ChatRequest` struct
which allows calls to specify a list of stop sequences
2024-03-01 12:08:11 -05:00
drbh 3dd7da2198
feat: accept legacy request format and response (#1527)
This WIP PR (will) add support for legacy OpenAI `v1/completions` API.

This should allow TGI to be a drop in replacement for OpenAI when using
tools that rely on the completions api

Should fix:
https://github.com/huggingface/text-generation-inference/issues/1468
2024-02-29 10:44:20 -05:00
Hugo Abonizio 9ed4d2c780
Fix async client timeout (#1617)
# What does this PR do?

Fixes #1616

According to the [aiohttp.ClientTimeout
docs](https://docs.aiohttp.org/en/stable/client_reference.html#aiohttp.ClientTimeout),
the arguments should be in seconds. This PR removes the multiplication
by 60.

@OlivierDehaene OR @Narsil
2024-02-29 15:41:49 +01:00
Nicolas Patry 5a3903ba99
Fix idefics default. (#1614)
# 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

 -->
2024-02-29 13:16:34 +01:00
drbh 343aa7a197
fix: Handle concurrent grammar requests (#1610)
This PR fixes parallel grammar requests, currently grammar states are
not concatenated correctly when a new request is added to the batch and
this results in incorrect generation. This PR updates the `concatenate`
function to correctly include the previous states.

fixes: #1601
2024-02-29 11:17:42 +01:00
OlivierDehaene e6bb3ff81f
v1.4.3 (#1609) 2024-02-28 16:12:14 +01:00
OlivierDehaene 26cdea5c0c
feat: Qwen2 (#1608)
See #1584

---------

Co-authored-by: Cheng Kuan Yong Jason <jasoncky96@gmail.com>
2024-02-28 15:50:31 +01:00
OlivierDehaene b40e833493
feat: starcoder2 (#1605) 2024-02-28 12:07:08 +01:00
Nicolas Patry 97e22369f4
Fixing guidance docs. (#1607)
# 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

 -->
2024-02-28 12:05:15 +01:00
Nicolas Patry 910d0a9062
Fixing x-compute-time. (#1606)
# What does this PR do?

It was meant to be in seconds float
<!--
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

 -->
2024-02-28 11:30:37 +01:00
drbh 9b6db5f793
Support tools (#1587)
This work in progress PR begins to add support for tools. Tools relies
on grammar support and still has some unsolved challenges. Opening the
PR for visibility and feedback
2024-02-28 11:10:27 +01:00
Nicolas Patry bf700e7eef
Revamp medusa implementation so that every model can benefit. (#1588)
# 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

 -->
2024-02-26 19:49:28 +01:00
drbh ac5a1c6f51
fix: avoid default message (#1579)
This PR avoids setting a default message in order to avoid unexpected
generations
2024-02-22 08:56:42 -05:00
OlivierDehaene 010508cec8
fix: fix openapi schema (#1586) 2024-02-21 15:30:45 +01:00
OlivierDehaene 9c1cb81cd8
v1.4.2 (#1585) 2024-02-21 14:50:57 +01:00
OlivierDehaene c86f58d37c
feat: add support for Gemma (#1583) 2024-02-21 14:15:22 +01:00
OlivierDehaene fa8a8e05af
fix(router): fix openapi and add jsonschema validation (#1578) 2024-02-21 11:05:32 +01:00
drbh c9f4c1af31
fix: refactor syntax to correctly include structs (#1580)
This PR fixes a compilation bug related to conditionally adding docs
behind a feature flag
2024-02-20 10:38:35 -05:00
drbh df23062574
improve endpoint support (#1577)
small PR to add a new interface endpoint behind a feature
2024-02-20 14:04:51 +01:00
Nicolas Patry d19c768cb8
Fix mistral with length > window_size for long prefills (rotary doesn't create long enough cos, sin). (#1571)
# 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

 -->
2024-02-19 15:23:12 +01:00
OlivierDehaene 4139054b82
v1.4.1 (#1568) 2024-02-16 17:50:57 +01:00
OlivierDehaene 0f2daad8b9
feat: add chat template struct to avoid tuple ordering errors (#1570) 2024-02-16 16:37:32 +01:00