Fix merge

.circleci/create_circleci_config.py

@@ -155,15 +155,15 @@ def to_dict(self):
                     elif self.name in ["flax","torch","tf"]:
                         name = self.name if self.name != "torch" else ""
                         if self.name == "torch":
-                            all_tests = glob.glob(f"tests/models/**/test_modeling_{name}*.py", recursive=True) 
+                            all_tests = glob.glob(f"tests/models/**/test_modeling_{name}*.py", recursive=True)
                             filtered = [k for k in all_tests if ("_tf_") not in k and "_flax_" not in k]
                             expanded_tests.extend(filtered)
                         else:
                             expanded_tests.extend(glob.glob(f"tests/models/**/test_modeling_{name}*.py", recursive=True))
                     else:
                         expanded_tests.extend(glob.glob("tests/models/**/test_modeling*.py", recursive=True))
                 elif test == "tests/pipelines":
-                    expanded_tests.extend(glob.glob("tests/models/**/test_modeling*.py", recursive=True)) 
+                    expanded_tests.extend(glob.glob("tests/models/**/test_modeling*.py", recursive=True))
                 else:
                     expanded_tests.append(test)
             tests = " ".join(expanded_tests)
@@ -326,7 +326,7 @@ def job_name(self):
     "examples_tensorflow",
     cache_name="tensorflow_examples",
     docker_image=[{"image":"huggingface/transformers-examples-tf"}],
-    install_steps=["uv venv && uv pip install ."],
+    install_steps=["uv venv && uv pip install . && uv pip install -r examples/tensorflow/_tests_requirements.txt"],
     parallelism=8
 )
 

.github/ISSUE_TEMPLATE/bug-report.yml

@@ -25,7 +25,7 @@ body:
 
         Models:
 
-          - text models: @ArthurZucker and @younesbelkada
+          - text models: @ArthurZucker 
           - vision models: @amyeroberts
           - speech models: @sanchit-gandhi
           - graph models: @clefourrier
@@ -44,7 +44,7 @@ body:
           - deepspeed: HF Trainer/Accelerate: @muellerzr
           - ray/raytune: @richardliaw, @amogkam
           - Big Model Inference: @SunMarc
-          - quantization (bitsandbytes, autogpt): @SunMarc and @younesbelkada
+          - quantization (bitsandbytes, autogpt): @SunMarc
 
         Documentation: @stevhliu
 

.github/PULL_REQUEST_TEMPLATE.md

@@ -39,7 +39,7 @@ members/contributors who may be interested in your PR.
 
 Models:
 
-- text models: @ArthurZucker and @younesbelkada
+- text models: @ArthurZucker
 - vision models: @amyeroberts
 - speech models: @sanchit-gandhi
 - graph models: @clefourrier
@@ -58,7 +58,7 @@ Integrations:
 - deepspeed: HF Trainer/Accelerate: @muellerzr
 - ray/raytune: @richardliaw, @amogkam
 - Big Model Inference: @SunMarc
-- quantization (bitsandbytes, autogpt): @SunMarc and @younesbelkada
+- quantization (bitsandbytes, autogpt): @SunMarc 
 
 Documentation: @stevhliu and @MKhalusova
 

.github/workflows/benchmark.yml

@@ -0,0 +1,42 @@
+name: Self-hosted runner (benchmark)
+
+on:
+  schedule:
+    - cron: "17 2 * * *"
+  workflow_call:
+
+env:
+  HF_HOME: /mnt/cache
+  TF_FORCE_GPU_ALLOW_GROWTH: true
+
+
+jobs:
+  benchmark:
+    name: Benchmark
+    runs-on: [single-gpu, nvidia-gpu, a10, ci]
+    container:
+      image: huggingface/transformers-all-latest-gpu
+      options: --gpus all --privileged --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+    steps:
+      - name: Update clone
+        working-directory: /transformers
+        run: |
+          git fetch && git checkout ${{ github.sha }}
+
+      - name: Reinstall transformers in edit mode (remove the one installed during docker image build)
+        working-directory: /transformers
+        run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
+
+      - name: Benchmark (daily)
+        if: github.event_name == 'schedule'
+        working-directory: /transformers
+        run: |
+          python3 -m pip install optimum-benchmark>=0.2.0
+          HF_TOKEN=${{ secrets.TRANSFORMERS_BENCHMARK_TOKEN }} python3 benchmark/benchmark.py --repo_id hf-internal-testing/benchmark_results --path_in_repo $(date +'%Y-%m-%d') --config-dir benchmark/config --config-name generation --commit=${{ github.sha }} backend.model=google/gemma-2b backend.cache_implementation=null,static backend.torch_compile=false,true --multirun
+
+      - name: Benchmark (merged to main event)
+        if: github.event_name == 'push' && github.ref_name == 'main'
+        working-directory: /transformers
+        run: |
+          python3 -m pip install optimum-benchmark>=0.2.0
+          HF_TOKEN=${{ secrets.TRANSFORMERS_BENCHMARK_TOKEN }} python3 benchmark/benchmark.py --repo_id hf-internal-testing/benchmark_results_merge_event --path_in_repo $(date +'%Y-%m-%d') --config-dir benchmark/config --config-name generation --commit=${{ github.sha }} backend.model=google/gemma-2b backend.cache_implementation=null,static backend.torch_compile=false,true --multirun

.github/workflows/build-nightly-ci-docker-images.yml

@@ -15,16 +15,6 @@ jobs:
     name: "Nightly PyTorch + Stable TensorFlow"
     runs-on: [intel-cpu, 8-cpu, ci]
     steps:
-      - name: Cleanup disk
-        run: |
-          sudo ls -l /usr/local/lib/
-          sudo ls -l /usr/share/
-          sudo du -sh /usr/local/lib/
-          sudo du -sh /usr/share/
-          sudo rm -rf /usr/local/lib/android
-          sudo rm -rf /usr/share/dotnet
-          sudo du -sh /usr/local/lib/
-          sudo du -sh /usr/share/
       -
         name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v2
@@ -52,16 +42,6 @@ jobs:
     name: "Nightly PyTorch + DeepSpeed"
     runs-on: [intel-cpu, 8-cpu, ci]
     steps:
-      - name: Cleanup disk
-        run: |
-          sudo ls -l /usr/local/lib/
-          sudo ls -l /usr/share/
-          sudo du -sh /usr/local/lib/
-          sudo du -sh /usr/share/
-          sudo rm -rf /usr/local/lib/android
-          sudo rm -rf /usr/share/dotnet
-          sudo du -sh /usr/local/lib/
-          sudo du -sh /usr/share/
       -
         name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v2

.github/workflows/model_jobs.yml

@@ -12,6 +12,12 @@ on:
       slice_id:
         required: true
         type: number
+      runner:
+        required: true
+        type: string
+      docker:
+        required: true
+        type: string
 
 env:
   HF_HOME: /mnt/cache
@@ -31,12 +37,13 @@ jobs:
   run_models_gpu:
     name: " "
     strategy:
+      max-parallel: 8
       fail-fast: false
       matrix:
         folders: ${{ fromJson(inputs.folder_slices)[inputs.slice_id] }}
-    runs-on: ['${{ inputs.machine_type }}', nvidia-gpu, t4, daily-ci]
+    runs-on: ['${{ inputs.machine_type }}', nvidia-gpu, t4, '${{ inputs.runner }}']
     container:
-      image: huggingface/transformers-all-latest-gpu
+      image: ${{ inputs.docker }}
       options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     steps:
       - name: Echo input and matrix info
@@ -65,6 +72,18 @@ jobs:
         working-directory: /transformers
         run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
 
+      - name: Update / Install some packages (for Past CI)
+        if: ${{ contains(inputs.docker, '-past-') }}
+        working-directory: /transformers
+        run: |
+          python3 -m pip install -U datasets
+
+      - name: Update / Install some packages (for Past CI)
+        if: ${{ contains(inputs.docker, '-past-') && contains(inputs.docker, '-pytorch-') }}
+        working-directory: /transformers
+        run: |
+          python3 -m pip install --no-cache-dir git+https://github.com/huggingface/accelerate@main#egg=accelerate
+
       - name: NVIDIA-SMI
         run: |
           nvidia-smi

.github/workflows/push-important-models.yml

@@ -133,3 +133,10 @@ jobs:
           slackChannel: ${{ secrets.SLACK_CIFEEDBACK_CHANNEL }}
           slackToken: ${{ secrets.SLACK_CIFEEDBACK_BOT_TOKEN }}
           waitForSSH: true
+
+  benchmark:
+    name: Benchmark workflow
+    needs: get_modified_models
+    if: ${{ needs.get_modified_models.outputs.matrix != '[]' && needs.get_modified_models.outputs.matrix != '' && fromJson(needs.get_modified_models.outputs.matrix)[0] != null }}
+    uses: ./.github/workflows/benchmark.yml
+    secrets: inherit

.github/workflows/self-nightly-caller.yml

@@ -0,0 +1,43 @@
+name: Self-hosted runner (nightly-ci)
+
+
+on:
+  repository_dispatch:
+  schedule:
+    - cron: "17 2 * * *"
+  push:
+    branches:
+      - run_nightly_ci*
+
+jobs:
+  build_nightly_ci_images:
+    name: Build Nightly CI Docker Images
+    if: (github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_nightly_ci'))
+    uses: ./.github/workflows/build-nightly-ci-docker-images.yml
+    secrets: inherit
+
+  model-ci:
+    name: Model CI
+    needs: [build_nightly_ci_images]
+    uses: ./.github/workflows/self-scheduled.yml
+    with:
+      job: run_models_gpu
+      slack_report_channel: "#transformers-ci-past-future"
+      runner: ci
+      docker: huggingface/transformers-all-latest-torch-nightly-gpu
+      ci_event: Nightly CI
+    secrets: inherit
+
+  deepspeed-ci:
+    name: DeepSpeed CI
+    needs: [build_nightly_ci_images]
+    uses: ./.github/workflows/self-scheduled.yml
+    with:
+      job: run_torch_cuda_extensions_gpu
+      slack_report_channel: "#transformers-ci-past-future"
+      runner: ci
+      # test deepspeed nightly build with the latest release torch
+      docker: huggingface/transformers-pytorch-deepspeed-latest-gpu
+      ci_event: Nightly CI
+      working-directory-prefix: /workspace
+    secrets: inherit