Merge branch 'pytorch:master' into master

Author: louis-she

.circleci/config.yml

@@ -4,11 +4,11 @@ parameters:
   pytorch_stable_image:
     type: string
     # https://hub.docker.com/r/pytorch/pytorch/tags
-    default: "pytorch/pytorch:1.11.0-cuda11.3-cudnn8-runtime"
+    default: "pytorch/pytorch:1.12.1-cuda11.3-cudnn8-runtime"
   pytorch_stable_image_devel:
     type: string
     # https://hub.docker.com/r/pytorch/pytorch/tags
-    default: "pytorch/pytorch:1.11.0-cuda11.3-cudnn8-devel"
+    default: "pytorch/pytorch:1.12.1-cuda11.3-cudnn8-devel"
   workingdir:
     type: string
     default: "/tmp/ignite"

.github/workflows/gpu-tests.yml

@@ -9,6 +9,7 @@ on:
       - "examples/**.py"
       - "requirements-dev.txt"
       - ".github/workflows/gpu-tests.yml"
+  workflow_dispatch:
 
 concurrency:
   # <workflow_name>-<branch_name>-<true || commit_sha (if branch is protected)>

.github/workflows/pytorch-version-tests.yml

@@ -5,6 +5,7 @@ on:
   schedule:
     # Run at 00:00 UTC Every Day
     - cron: "0 0 * * *"
+  workflow_dispatch:
 
 jobs:
   build:
@@ -14,7 +15,8 @@ jobs:
       fail-fast: false
       matrix:
         python-version: [3.7, 3.8, 3.9]
-        pytorch-version: [1.9.1, 1.8.1, 1.7.1, 1.6.0, 1.5.1, 1.4.0, 1.3.1]
+        pytorch-version:
+          [1.11.0, 1.10.0, 1.9.1, 1.8.1, 1.7.1, 1.6.0, 1.5.1, 1.4.0, 1.3.1]
         exclude:
           - pytorch-version: 1.3.1
             python-version: 3.8
@@ -23,9 +25,9 @@ jobs:
           - pytorch-version: 1.4.0
             python-version: 3.9
           - pytorch-version: 1.5.1
-            python-version: 3.9 
+            python-version: 3.9
           - pytorch-version: 1.6.0
-            python-version: 3.9 
+            python-version: 3.9
           # disabling python 3.9 support with PyTorch 1.7.1 and 1.8.1, to stop repeated pytorch-version test fail.
           # https://github.com/pytorch/ignite/issues/2383
           - pytorch-version: 1.7.1
@@ -72,7 +74,7 @@ jobs:
           python setup.py install
 
       - name: Install appropriate Pillow for PyTorch 1.3.1
-        shell: bash -l {0}      
+        shell: bash -l {0}
         if: ${{ matrix.pytorch-version == '1.3.1' }}
         run: |
           pip install --upgrade 'Pillow<7'
@@ -86,7 +88,7 @@ jobs:
       - name: Run Tests
         shell: bash -l {0}
         run: |
-          pytest tests -vvv
+          bash tests/run_cpu_tests.sh "not test_time_profilers"
 
   create-issue:
     runs-on: ubuntu-latest

.github/workflows/unit-tests.yml

@@ -49,10 +49,6 @@ jobs:
             pytorch-channel: pytorch
             python-version: 3.8
             skip-distrib-tests: 1
-        exclude:
-          # temporarily exclude stable on 3.10 as not yet released
-          - pytorch-channel: pytorch
-            python-version: "3.10"
 
     steps:
       - uses: actions/checkout@v2
@@ -100,7 +96,7 @@ jobs:
           bash ./tests/run_code_style.sh lint
 
       - name: Run Mypy
-        if: ${{ matrix.os == 'ubuntu-latest' }}
+        if: ${{ matrix.os == 'ubuntu-latest' && matrix.pytorch-channel == 'pytorch-nightly'}}
         run: |
           bash ./tests/run_code_style.sh mypy
 

docker/docker.cfg

@@ -1,4 +1,4 @@
 [DEFAULT]
-build_docker_image_pytorch_version = 1.11.0-cuda11.3-cudnn8
-build_docker_image_hvd_version = v0.24.2
-build_docker_image_msdp_version = v0.6.0
+build_docker_image_pytorch_version = 1.12.1-cuda11.3-cudnn8
+build_docker_image_hvd_version = v0.25.0
+build_docker_image_msdp_version = v0.7.2

docker/hvd/Dockerfile.hvd-apex

@@ -9,6 +9,12 @@ FROM pytorch/pytorch:${PTH_VERSION}-devel AS apex-hvd-builder
 ARG ARG_TORCH_CUDA_ARCH_LIST="3.5 5.2 6.0 6.1 7.0+PTX 8.0"
 ENV TORCH_CUDA_ARCH_LIST=$ARG_TORCH_CUDA_ARCH_LIST
 
+# Renew nvidia signing key
+# https://developer.nvidia.com/blog/updating-the-cuda-linux-gpg-repository-key/
+RUN apt-key del 7fa2af80 && \
+    rm /etc/apt/sources.list.d/nvidia-ml.list && \
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
+
 # Install git
 RUN apt-get update && apt-get install -y --no-install-recommends git && \
      rm -rf /var/lib/apt/lists/*

docker/hvd/Dockerfile.hvd-base

@@ -7,6 +7,12 @@ FROM pytorch/pytorch:${PTH_VERSION}-devel as builder
 
 ARG HVD_VERSION
 
+# Renew nvidia signing key
+# https://developer.nvidia.com/blog/updating-the-cuda-linux-gpg-repository-key/
+RUN apt-key del 7fa2af80 && \
+    rm /etc/apt/sources.list.d/nvidia-ml.list && \
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
+
 # Build Horovod
 RUN apt-get update && apt-get install -y git && \
         git clone --recursive --depth 1 --branch ${HVD_VERSION} https://github.com/horovod/horovod.git /horovod && \

docker/main/Dockerfile.apex

@@ -9,6 +9,12 @@ FROM pytorch/pytorch:${PTH_VERSION}-devel AS apex-builder
 ARG ARG_TORCH_CUDA_ARCH_LIST="3.5 5.2 6.0 6.1 7.0+PTX 8.0"
 ENV TORCH_CUDA_ARCH_LIST=$ARG_TORCH_CUDA_ARCH_LIST
 
+# Renew nvidia signing key
+# https://developer.nvidia.com/blog/updating-the-cuda-linux-gpg-repository-key/
+RUN apt-key del 7fa2af80 && \
+    rm /etc/apt/sources.list.d/nvidia-ml.list && \
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
+
 # Install git
 RUN apt-get update && apt-get install -y --no-install-recommends git && \
      rm -rf /var/lib/apt/lists/*

docker/msdp/Dockerfile.msdp-apex

@@ -9,6 +9,12 @@ FROM pytorch/pytorch:${PTH_VERSION}-devel AS apex-msdp-builder
 ARG ARG_TORCH_CUDA_ARCH_LIST="3.5 5.2 6.0 6.1 7.0+PTX 8.0"
 ENV TORCH_CUDA_ARCH_LIST=$ARG_TORCH_CUDA_ARCH_LIST
 
+# Renew nvidia signing key
+# https://developer.nvidia.com/blog/updating-the-cuda-linux-gpg-repository-key/
+RUN apt-key del 7fa2af80 && \
+    rm /etc/apt/sources.list.d/nvidia-ml.list && \
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
+
 # Install git
 RUN apt-get update && apt-get install -y --no-install-recommends git && \
      rm -rf /var/lib/apt/lists/*
@@ -50,6 +56,12 @@ RUN cd /msdp && export CUDA_HOME=/usr/local/cuda && \
     pip install --no-cache-dir deepspeed-*.whl && \
     rm -fr /msdp
 
+# Renew nvidia signing key
+# https://developer.nvidia.com/blog/updating-the-cuda-linux-gpg-repository-key/
+RUN apt-key del 7fa2af80 && \
+    rm /etc/apt/sources.list.d/nvidia-ml.list && \
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
+
 # Install tzdata / git
 RUN apt-get update && \
     ln -fs /usr/share/zoneinfo/Europe/Paris /etc/localtime && \

docker/test_image.py

@@ -23,6 +23,9 @@
 def run_python_cmd(cmd):
 
     try_except_cmd = f"""
+import warnings
+warnings.filterwarnings("ignore")
+
 def main():
     {cmd}