1693 update for pytorch 1.8

.github/workflows/cron.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: [self-hosted, linux, x64, common]
     strategy:
       matrix:
-        pytorch-version: [1.5.0, 1.5.1, 1.6.0, latest]
+        pytorch-version: [1.5.1, 1.6.0, 1.7.1, latest]
     steps:
     - uses: actions/checkout@v2
     - name: Install the dependencies
@@ -23,15 +23,12 @@ jobs:
         python -m pip uninstall -y torch torchvision
         if [ ${{ matrix.pytorch-version }} == "latest" ]; then
           python -m pip install torch torchvision
-        elif [ ${{ matrix.pytorch-version }} == "1.5.0" ]; then
-          python -m pip install torch==1.5.0
-          python -m pip install torchvision==0.6.0
         elif [ ${{ matrix.pytorch-version }} == "1.5.1" ]; then
-          python -m pip install torch==1.5.1
-          python -m pip install torchvision==0.6.1
+          python -m pip install torch==1.5.1 torchvision==0.6.1
         elif [ ${{ matrix.pytorch-version }} == "1.6.0" ]; then
-          python -m pip install torch==1.6.0
-          python -m pip install torchvision==0.7.0
+          python -m pip install torch==1.6.0 torchvision==0.7.0
+        elif [ ${{ matrix.pytorch-version }} == "1.7.1" ]; then
+          python -m pip install torch==1.7.1 torchvision==0.8.2
         fi
         python -m pip install -r requirements-dev.txt
         python -m pip list

.github/workflows/integration.yml

@@ -7,7 +7,7 @@ on:
 jobs:
   integration-py3:
     container:
-      image: nvcr.io/nvidia/pytorch:20.03-py3  # CUDA 10.2
+      image: nvcr.io/nvidia/pytorch:20.12-py3  # CUDA 11.1, default for PT 1.8.0
       options: --gpus all
     runs-on: [self-hosted, linux, x64, common]
     steps:
@@ -28,13 +28,13 @@ jobs:
         path: |
           ~/.cache/pip
           ~/.cache/torch
-        key: docker-20-03-py3-pip-${{ steps.pip-cache.outputs.datew }}
+        key: docker-py3-pip-${{ steps.pip-cache.outputs.datew }}
     - name: Install the dependencies
       run: |
         which python
         python -m pip install --upgrade pip wheel
         python -m pip uninstall -y torch torchvision
-        python -m pip install torch==1.7.1 torchvision==0.8.2
+        python -m pip install torch==1.8.0 torchvision==0.9.0
         python -m pip install -r requirements-dev.txt
     - name: Run integration tests
       run: |

.github/workflows/pythonapp.yml

@@ -9,7 +9,7 @@ on:
 
 jobs:
   # caching of these jobs:
-  #   - docker-20-03-py3-pip- (shared)
+  #   - docker-py3-pip- (shared)
   #   - ubuntu py37 pip-
   #   - os-latest-pip- (shared)
   flake8-py3:
@@ -80,13 +80,12 @@ jobs:
     - if: runner.os == 'windows'
       name: Install torch cpu from pytorch.org (Windows only)
       run: |
-        python -m pip install torch==1.7.1+cpu torchvision==0.8.2+cpu -f https://download.pytorch.org/whl/torch_stable.html
+        python -m pip install torch==1.8.0+cpu torchvision==0.9.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
         # min. requirements for windows instances
         python -c "f=open('requirements-dev.txt', 'r'); txt=f.readlines(); f.close(); print(txt); f=open('requirements-dev.txt', 'w'); f.writelines(txt[1:12]); f.close()"
     - name: Install the dependencies
       run: |
-        python -m pip install torch==1.7.1
-        python -m pip install torchvision==0.8.2
+        python -m pip install torch==1.8.0 torchvision==0.9.0
         cat "requirements-dev.txt"
         python -m pip install -r requirements-dev.txt
         python -m pip list
@@ -134,11 +133,11 @@ jobs:
     - if: runner.os == 'windows'
       name: Install torch cpu from pytorch.org (Windows only)
       run: |
-        python -m pip install torch==1.7.1+cpu -f https://download.pytorch.org/whl/torch_stable.html
+        python -m pip install torch==1.8.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
     - name: Install the dependencies
       run: |
         # min. requirements
-        python -m pip install torch==1.7.1
+        python -m pip install torch==1.8.0
         python -m pip install -r requirements-min.txt
         python -m pip list
         BUILD_MONAI=0 python setup.py develop  # no compile of extensions
@@ -156,15 +155,13 @@ jobs:
     strategy:
       matrix:
         environment:
-          - "PT15+CUDA101"
           - "PT16+CUDA102"
           - "PT16+CUDA110"
           - "PT17+CUDA102"
           - "PT17+CUDA110"
+          - "PT18+CUDA102"
+          - "PT18+CUDA112"
         include:
-          - environment: PT15+CUDA101
-            pytorch: "torch==1.5.0+cu101 torchvision==0.6.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html"
-            base: "nvcr.io/nvidia/cuda:10.1-devel-ubuntu18.04"
           - environment: PT16+CUDA102
             pytorch: "torch==1.6.0 torchvision==0.7.0"
             base: "nvcr.io/nvidia/cuda:10.2-devel-ubuntu18.04"
@@ -179,6 +176,13 @@ jobs:
             # we explicitly set pytorch to -h to avoid pip install error
             pytorch: "-h"
             base: "nvcr.io/nvidia/pytorch:20.09-py3"
+          - environment: PT18+CUDA102
+            pytorch: "torch==1.8.0 torchvision==0.9.0"
+            base: "nvcr.io/nvidia/cuda:10.2-devel-ubuntu18.04"
+          - environment: PT18+CUDA112
+            # we explicitly set pytorch to -h to avoid pip install error
+            pytorch: "-h"
+            base: "nvcr.io/nvidia/pytorch:21.02-py3"
     container:
       image: ${{ matrix.base }}
       options: --gpus all
@@ -187,7 +191,10 @@ jobs:
     - uses: actions/checkout@v2
     - name: apt install
       run: |
-        if [ ${{ matrix.environment }} != "PT16+CUDA110" ]; then \
+        if [ ${{ matrix.environment }} = "PT16+CUDA102" ] || \
+           [ ${{ matrix.environment }} = "PT17+CUDA102" ] || \
+           [ ${{ matrix.environment }} = "PT18+CUDA102" ]
+        then
         PYVER=3.6 PYSFX=3 DISTUTILS=python3-distutils && \
         apt-get update && apt-get install -y --no-install-recommends \
           curl \
@@ -217,7 +224,8 @@ jobs:
         ln -s /usr/bin/python$PYVER /usr/bin/python`echo $PYVER | cut -c1-1` &&
         curl -O https://bootstrap.pypa.io/get-pip.py && \
         python get-pip.py && \
-        rm get-pip.py ; fi
+        rm get-pip.py;
+        fi
     - name: Install dependencies
       run: |
         which python
@@ -231,10 +239,10 @@ jobs:
         export CUDA_VISIBLE_DEVICES=$(coverage run -m tests.utils)
         echo $CUDA_VISIBLE_DEVICES
         python -c "import torch; print(torch.__version__); print('{} of GPUs available'.format(torch.cuda.device_count()))"
-        python -c 'import torch; print(torch.rand(5,3, device=torch.device("cuda:0")))'
+        python -c 'import torch; print(torch.rand(5, 3, device=torch.device("cuda:0")))'
         python -c "import monai; monai.config.print_config()"
         BUILD_MONAI=1 ./runtests.sh --quick
-        if [ ${{ matrix.environment }} == "PT16+CUDA110" ]; then
+        if [ ${{ matrix.environment }} == "PT18+CUDA112" ]; then
           # test the clang-format tool downloading once
           coverage run -m tests.clang_format_utils
         fi

.github/workflows/setupapp.yml

@@ -8,7 +8,7 @@ on:
 
 jobs:
   # caching of these jobs:
-  #   - docker-20-03-py3-pip- (shared)
+  #   - docker-py3-pip- (shared)
   #   - ubuntu py36 37 38-pip-
   #   - os-latest-pip (shared)
   coverage-py3:
@@ -30,7 +30,7 @@ jobs:
         path: |
           ~/.cache/pip
           ~/.cache/torch
-        key: docker-20-03-py3-pip-${{ steps.pip-cache.outputs.datew }}
+        key: docker-py3-pip-${{ steps.pip-cache.outputs.datew }}
     - name: Install the dependencies
       run: |
         which python
@@ -51,7 +51,7 @@ jobs:
     - name: Upload coverage
       uses: codecov/codecov-action@v1
       with:
-        fail_ci_if_error: true
+        fail_ci_if_error: false
         file: ./coverage.xml
 
   test-py3x:

tests/utils.py

@@ -569,13 +569,13 @@ def query_memory(n=2):
     """
     Find best n idle devices and return a string of device ids.
     """
-    bash_string = "nvidia-smi --query-gpu=utilization.gpu,temperature.gpu,memory.used --format=csv,noheader,nounits"
+    bash_string = "nvidia-smi --query-gpu=utilization.gpu,power.draw,memory.used --format=csv,noheader,nounits"
 
     try:
         p1 = Popen(bash_string.split(), stdout=PIPE)
         output, error = p1.communicate()
         free_memory = [x.split(",") for x in output.decode("utf-8").split("\n")[:-1]]
-        free_memory = np.asarray(free_memory, dtype=np.float).T
+        free_memory = np.asarray(free_memory, dtype=float).T
         ids = np.lexsort(free_memory)[:n]
     except (FileNotFoundError, TypeError, IndexError):
         ids = range(n) if isinstance(n, int) else []