Merge branch 'main' into bnb

.circleci/create_circleci_config.py

@@ -133,7 +133,7 @@ def to_dict(self):
                 "command": """dpkg-query --show --showformat='${Installed-Size}\t${Package}\n' | sort -rh | head -25 | sort -h | awk '{ package=$2; sub(".*/", "", package); printf("%.5f GB %s\n", $1/1024/1024, package)}' || true"""}
             },
             {"run": {"name": "Create `test-results` directory", "command": "mkdir test-results"}},
-            {"run": {"name": "Get files to test", "command":f'curl -L -o {self.job_name}_test_list.txt <<pipeline.parameters.{self.job_name}_test_list>>' if self.name != "pr_documentation_tests" else 'echo "Skipped"'}},
+            {"run": {"name": "Get files to test", "command":f'curl -L -o {self.job_name}_test_list.txt <<pipeline.parameters.{self.job_name}_test_list>> --header "Circle-Token: $CIRCLE_TOKEN"' if self.name != "pr_documentation_tests" else 'echo "Skipped"'}},
                         {"run": {"name": "Split tests across parallel nodes: show current parallel tests",
                     "command": f"TESTS=$(circleci tests split  --split-by=timings {self.job_name}_test_list.txt) && echo $TESTS > splitted_tests.txt && echo $TESTS | tr ' ' '\n'" if self.parallelism else f"awk '{{printf \"%s \", $0}}' {self.job_name}_test_list.txt > splitted_tests.txt"
                     }
@@ -371,9 +371,14 @@ def create_circleci_config(folder=None):
             **{j.job_name + "_test_list":{"type":"string", "default":''} for j in jobs},
             **{j.job_name + "_parallelism":{"type":"integer", "default":1} for j in jobs},
         },
-        "jobs" : {j.job_name: j.to_dict() for j in jobs},
-        "workflows": {"version": 2, "run_tests": {"jobs": [j.job_name for j in jobs]}}
+        "jobs" : {j.job_name: j.to_dict() for j in jobs}
     }
+    if "CIRCLE_TOKEN" in os.environ:
+        # For private forked repo. (e.g. new model addition)
+        config["workflows"] = {"version": 2, "run_tests": {"jobs": [{j.job_name: {"context": ["TRANSFORMERS_CONTEXT"]}} for j in jobs]}}
+    else:
+        # For public repo. (e.g. `transformers`)
+        config["workflows"] = {"version": 2, "run_tests": {"jobs": [j.job_name for j in jobs]}}
     with open(os.path.join(folder, "generated_config.yml"), "w") as f:
         f.write(yaml.dump(config, sort_keys=False, default_flow_style=False).replace("' << pipeline", " << pipeline").replace(">> '", " >>"))
 

docker/transformers-quantization-latest-gpu/Dockerfile

@@ -9,7 +9,7 @@ SHELL ["sh", "-lc"]
 # The following `ARG` are mainly used to specify the versions explicitly & directly in this docker file, and not meant
 # to be used as arguments for docker build (so far).
 
-ARG PYTORCH='2.4.1'
+ARG PYTORCH='2.5.1'
 # Example: `cu102`, `cu113`, etc.
 ARG CUDA='cu118'
 
@@ -36,12 +36,17 @@ RUN python3 -m pip install --no-cache-dir einops
 # Add bitsandbytes for mixed int8 testing
 RUN python3 -m pip install --no-cache-dir bitsandbytes
 
-# Add auto-gptq for gtpq quantization testing
-RUN python3 -m pip install --no-cache-dir auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/
+# Add auto-gptq for gtpq quantization testing, installed from source for pytorch==2.5.1 compatibility
+# TORCH_CUDA_ARCH_LIST="7.5+PTX" is added to make the package compile for Tesla T4 gpus available for the CI.
+RUN pip install gekko
+RUN git clone https://github.com/PanQiWei/AutoGPTQ.git && cd AutoGPTQ && TORCH_CUDA_ARCH_LIST="7.5+PTX" python3 setup.py install
 
 # Add optimum for gptq quantization testing
 RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/optimum@main#egg=optimum
 
+# Add PEFT
+RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/peft@main#egg=peft
+
 # Add aqlm for quantization testing
 RUN python3 -m pip install --no-cache-dir aqlm[gpu]==1.0.2
 
@@ -52,8 +57,8 @@ RUN python3 -m pip install --no-cache-dir hqq
 RUN python3 -m pip install --no-cache-dir gguf
 
 # Add autoawq for quantization testing
-# >=v0.2.3 needed for compatibility with torch 2.2.1
-RUN python3 -m pip install --no-cache-dir https://github.com/casper-hansen/AutoAWQ/releases/download/v0.2.3/autoawq-0.2.3+cu118-cp310-cp310-linux_x86_64.whl
+# >=v0.2.7 needed for compatibility with transformers > 4.46
+RUN python3 -m pip install --no-cache-dir https://github.com/casper-hansen/AutoAWQ/releases/download/v0.2.7.post2/autoawq-0.2.7.post2-py3-none-any.whl
 
 # Add quanto for quantization testing
 RUN python3 -m pip install --no-cache-dir optimum-quanto

docs/source/ar/_toctree.yml

@@ -129,10 +129,10 @@
     title: التصدير إلى TFLite
   - local: torchscript
     title: التصدير إلى TorchScript
-#   - local: benchmarks
-#     title: المعايير
-#   - local: notebooks
-#     title: دفاتر الملاحظات مع الأمثلة
+  - local: benchmarks
+    title: المعايير
+  - local: notebooks
+    title: دفاتر الملاحظات مع الأمثلة
 #   - local: community
 #     title: موارد المجتمع
   - local: troubleshooting