1. 26 Jan, 2024 5 commits
    • Nicolas Patry's avatar
      Fixing top_n_tokens. (#1497) · 069895b9
      Nicolas Patry authored
      # What does this PR do?
      
      Superseeds #1459
      
      The fix works as follows.
      We updated next_token_chooser to return all logprbs, then
      batch_top_n_tokens, now also gets accepted_ids + speculated_length (so
      it knows how to interpret the flat logprobs).
      
      We then update the code to return lists ot `Tokens` that it expects.
      <!--
      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
      
       -->
      069895b9
    • OlivierDehaene's avatar
      v1.4.0 (#1494) · c2d4a3b5
      OlivierDehaene authored
      c2d4a3b5
    • fxmarty's avatar
      GPTQ support on ROCm (#1489) · 650fea18
      fxmarty authored
      
      
      Tested with
      ```
      CUDA_VISIBLE_DEVICES=0 text-generation-launcher --model-id TheBloke/Llama-2-7B-Chat-GPTQ --quantize gptq
      EXLLAMA_VERSION=1 CUDA_VISIBLE_DEVICES=0 text-generation-launcher --model-id TheBloke/Llama-2-7B-Chat-GPTQ --quantize gptq
      CUDA_VISIBLE_DEVICES="0,1" text-generation-launcher --model-id TheBloke/Llama-2-7B-Chat-GPTQ --quantize gptq
      ```
      
      all with good and identical results on MI210.
      
      ---------
      Co-authored-by: default avatarFelix Marty <felix@hf.co>
      Co-authored-by: default avatarOlivierDehaene <olivier@huggingface.co>
      Co-authored-by: default avatarOlivierDehaene <23298448+OlivierDehaene@users.noreply.github.com>
      650fea18
    • Nicolas Patry's avatar
      Add sealion mpt support (#1477) · ac499727
      Nicolas Patry authored
      # 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: default avatarChoon Meng Tan <choonmeng@aisingapore.org>
      Co-authored-by: default avatarDavid Ong Tat-Wee <13075447+ongtw@users.noreply.github.com>
      ac499727
    • Nicolas Patry's avatar
      Reinstate exl2 with tp (#1490) · b9573218
      Nicolas Patry authored
      # 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
      
       -->
      b9573218
  2. 25 Jan, 2024 1 commit
    • drbh's avatar
      feat: adds phi model (#1442) · 7e2a7433
      drbh authored
      This PR adds basic modeling for phi-2 
      
      run
      ```bash
      text-generation-server \
          serve \
          microsoft/phi-2 \
          --revision 834565c23f9b28b96ccbeabe614dd906b6db551a
      ```
      
      
      test
      ```bash
      curl -s localhost:3000/generate \
         -X POST \
         -d '{"inputs":"What is Deep Learning?","parameters":{"max_new_tokens":20}}' \
         -H 'Content-Type: application/json' | jq .
      # {
      #   "generated_text": "\nDeep learning is a subset of machine learning that uses artificial neural networks to learn from data. These"
      # }
      ```
      
      
      
      notes 
      - recently (~1 day ago) the Phi weights and model were updated to
      accommodate adding [GQA/MQA attention to the
      model.](https://github.com/huggingface/transformers/pull/28163) This
      impl expects the original model format so a fixed revision is required
      at the moment.
      - this PR only includes a basic implementation of the model and can
      later be extended for support Flash and Sharded versions as well as make
      use of better optimization
      7e2a7433
  3. 24 Jan, 2024 1 commit
    • Nicolas Patry's avatar
      Fixing non divisible embeddings. (#1476) · 7e542d4d
      Nicolas Patry authored
      # 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
      
       -->
      7e542d4d
  4. 10 Jan, 2024 1 commit
    • PYNing's avatar
      Fix local load for Medusa (#1420) · da27fbdf
      PYNing authored
      # 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 -->
      
      Close #1418 
      Close #1415
      
      ## 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
      
       -->
      da27fbdf
  5. 09 Jan, 2024 2 commits
  6. 22 Dec, 2023 1 commit
  7. 21 Dec, 2023 2 commits
    • Nicolas Patry's avatar
      Fix local load for peft (#1373) · 529d7c25
      Nicolas Patry authored
      local directory overloaded still needs the directory to locate the
      weights files correctly.
      
      # 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
      
       -->
      529d7c25
    • OlivierDehaene's avatar
      564199ba
  8. 20 Dec, 2023 1 commit
    • Nicolas Patry's avatar
      Peft safetensors. (#1364) · eb8923a9
      Nicolas Patry authored
      Works by removing adapter_model.safetensors from being detected as the
      core model file (which skips the real peft detection).
      
      # 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
      
       -->
      eb8923a9
  9. 18 Dec, 2023 2 commits
  10. 15 Dec, 2023 4 commits
  11. 14 Dec, 2023 6 commits
  12. 12 Dec, 2023 2 commits
  13. 11 Dec, 2023 5 commits
  14. 30 Nov, 2023 1 commit
  15. 28 Nov, 2023 2 commits
    • Nicolas Patry's avatar
      Let each model resolve their own default dtype. (#1287) · ba552e1a
      Nicolas Patry authored
      # 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
      
       -->
      ba552e1a
    • Nicolas Patry's avatar
      `make install-flash-attn-v2-cuda` should work like `make... · 3c71c656
      Nicolas Patry authored
      `make install-flash-attn-v2-cuda` should work like `make install-flash-attn-v2` used to work. (#1294)
      
      # 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
      
       -->
      3c71c656
  16. 27 Nov, 2023 1 commit
    • fxmarty's avatar
      Add RoCm support (#1243) · b2b5df0e
      fxmarty authored
      
      
      This PR adds support for AMD Instinct MI210 & MI250 GPUs, with paged
      attention and FAv2 support.
      
      Remaining items to discuss, on top of possible others:
      * Should we have a
      `ghcr.io/huggingface/text-generation-inference:1.1.0+rocm` hosted image,
      or is it too early?
      * Should we set up a CI on MI210/MI250? I don't have access to the
      runners of TGI though.
      * Are we comfortable with those changes being directly in TGI, or do we
      need a fork?
      
      ---------
      Co-authored-by: default avatarFelix Marty <felix@hf.co>
      Co-authored-by: default avatarOlivierDehaene <olivier@huggingface.co>
      Co-authored-by: default avatarYour Name <you@example.com>
      b2b5df0e
  17. 25 Nov, 2023 1 commit
    • Nicolas Patry's avatar
      Exllama v2 (#1211) · ed2a3f61
      Nicolas Patry authored
      # What does this PR do?
      
      See #1165
      
      <!--
      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: default avatarFlorian Zimmermeister <flozi00.fz@gmail.com>
      Co-authored-by: default avatarUbuntu <ubuntu@ip-172-31-24-153.ec2.internal>
      ed2a3f61
  18. 23 Nov, 2023 2 commits
    • Vince Jankovics's avatar
      Fix IDEFICS dtype (#1214) · c6bb7670
      Vince Jankovics authored
      # 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 -->
      
      This forces the use of `bfloat16` for IDEFICS. The issue is that with
      `float16` the 80b model gives garbage output. Let me know if this
      solution is not appropriate and I'll adjust accordingly. For the details
      see below.
      
      The current behaviour:
      ```sh
      $ curl 127.0.0.1:8080/generate -X POST -d '{"inputs":"What is Deep Learning?","parameters":{"max_new_tokens":20}}' -H 'Content-Type: application/json'
      {"generated_text":""}
      ```
      
      On closer inspection with:
      ```python
      import requests
      
      headers = { "Content-Type": "application/json"}
      
      query = "What is Deep Learning?"
      data = {
          "inputs": query,
          "parameters": {
              "max_new_tokens": 10,
              "return_full_text": True,
              "decoder_input_details": True,
              "do_sample": False,
          },
      }
      
      api_url = "http://127.0.0.1:8080"
      response = requests.post(api_url + "/generate", headers=headers, json=data).json()
      
      for i in ['prefill', 'tokens']:
          print(f'### {i}')
          print(repr(''.join([t['text'] for t in response['details'][i]])))
      ```
      
      Prints:
      ```
      ### prefill
      '<s>WhatisDeepLearning?'
      ### tokens
      '<unk><unk><unk><unk><unk><unk><unk><unk><unk><unk>'
      ########
      ```
      
      With the change in this PR it prints:
      ```
      ### prefill
      '<s>WhatisDeepLearning?'
      ### tokens
      '\n\nDeep Learning is a subset of machine'
      ```
      
      Note, using the Transformers implementation (with
      `IdeficsForVisionText2Text.from_pretrained`) produces the latter
      (correct) output as well.
      This only happens with the 80b model, the 9b model is not as sensitive
      to the dtype (as also mentioned in the code).
      
      The reason for "forcing" this in the IDEFICS init method, is because if
      quantization is used, then the dtype cannot be set explicitly. And since
      it's left as `None`, it's set to `float16` by default
      [here](https://github.com/huggingface/text-generation-inference/blob/96a982ad8fc232479384476b1596a880697cc1d0/server/text_generation_server/models/__init__.py#L90).
      I.e. there's no other way to manually change the dtype if someone is
      using quantization:
      ```sh
      $ docker run .... ghcr.io/huggingface/text-generation-inference:latest --model-id HuggingFaceM4/idefics-80b-instruct --dtype bfloat16 --quantize bitsandbytes-nf4
      .....
      2023-10-31T12:42:26.710401Z  INFO shard-manager: text_generation_launcher: Starting shard rank=0
      2023-10-31T12:42:30.315734Z ERROR shard-manager: text_generation_launcher: Shard complete standard error output:
      
      Traceback (most recent call last):
      
        File "/opt/conda/bin/text-generation-server", line 8, in <module>
          sys.exit(app())
      
        File "/opt/conda/lib/python3.9/site-packages/text_generation_server/cli.py", line 80, in serve
          raise RuntimeError(
      
      RuntimeError: Only 1 can be set between `dtype` and `quantize`, as they both decide how goes the final model.
       rank=0
      Error: ShardCannotStart
      2023-10-31T12:42:30.414010Z ERROR text_generation_launcher: Shard 0 failed to start
      2023-10-31T12:42:30.414044Z  INFO text_generation_launcher: Shutting down shards
      ```
      
      ## Before submitting
      - [ ] This PR fixes a typo or improves the docs (you can dismiss the
      other checks if that's the case).
      - [x] 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.
      
      @Narsil what do you think?
      
      <!-- 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: default avatarNicolas Patry <patry.nicolas@protonmail.com>
      c6bb7670
    • OlivierDehaene's avatar
      chore: update to torch 2.1.0 (#1182) · 35509ff5
      OlivierDehaene authored
      Close #1142
      35509ff5