Print more library versions in CI (#17384)

* print more lib. versions and just befor test runs

* update print_env_pt.py

* rename to print_env

* Disable warning + better job name

* print python version
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

.github/workflows/doctests.yml

@@ -32,9 +32,7 @@ jobs:
 
       - name: GPU visibility
         run: |
-          utils/print_env_pt.py
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
+          python3 utils/print_env.py
 
       - name: Prepare files for doctests
         run: |

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

@@ -41,7 +41,7 @@ jobs:
 
             - name: Are GPUs recognized by our DL frameworks
               run: |
-                utils/print_env_pt.py
+                utils/print_env.py
 
             - name: Run all tests on GPU
               run: |
@@ -109,7 +109,7 @@ jobs:
 
             - name: Are GPUs recognized by our DL frameworks
               run: |
-                utils/print_env_pt.py
+                utils/print_env.py
 
             - name: Run all tests on GPU
               env:
@@ -163,7 +163,7 @@ jobs:
 
             - name: Are GPUs recognized by our DL frameworks
               run: |
-                utils/print_env_pt.py
+                utils/print_env.py
 
             - name: Run all tests on GPU
               run: |
@@ -206,7 +206,7 @@ jobs:
 
             - name: Are GPUs recognized by our DL frameworks
               run: |
-                utils/print_env_pt.py
+                utils/print_env.py
 
             - name: Run all tests on GPU
               run: |

.github/workflows/self-push.yml

@@ -87,17 +87,6 @@ jobs:
       image: huggingface/transformers-all-latest-gpu
       options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     steps:
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
-      - name: Are GPUs recognized by our DL frameworks
-        working-directory: /transformers
-        run: |
-          utils/print_env_pt.py
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
-
       - name: Echo folder ${{ matrix.folders }}
         shell: bash
         # For folders like `models/bert`, set an env. var. (`matrix_folders`) to `models_bert`, which will be used to
@@ -114,6 +103,15 @@ jobs:
         working-directory: /transformers
         run: git fetch && git checkout ${{ github.sha }}
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
       - name: Run all non-slow selected tests on GPU
         working-directory: /transformers
         run: |
@@ -146,17 +144,6 @@ jobs:
       image: huggingface/transformers-all-latest-gpu
       options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     steps:
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
-      - name: Are GPUs recognized by our DL frameworks
-        working-directory: /transformers
-        run: |
-          utils/print_env_pt.py
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
-
       - name: Echo folder ${{ matrix.folders }}
         shell: bash
         # For folders like `models/bert`, set an env. var. (`matrix_folders`) to `models_bert`, which will be used to
@@ -173,6 +160,15 @@ jobs:
         working-directory: /transformers
         run: git fetch && git checkout ${{ github.sha }}
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
       - name: Run all non-slow selected tests on GPU
         env:
           MKL_SERVICE_FORCE_INTEL: 1
@@ -210,19 +206,19 @@ jobs:
         with:
           fetch-depth: 2
 
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
       - name: Install dependencies
         run: |
           apt -y update && apt install -y libaio-dev
           pip install --upgrade pip
           pip install .[deepspeed-testing]
 
-      - name: Are GPUs recognized by our DL frameworks
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
         run: |
-          utils/print_env_pt.py
+          python utils/print_env.py
 
       - name: Run all non-slow selected tests on GPU
         # TODO: Here we pass all tests in the 2 folders for simplicity. It's better to pass only the identified tests.
@@ -259,10 +255,6 @@ jobs:
         with:
           fetch-depth: 2
 
-      - name: NVIDIA-SMI
-        run: |
-          nvidia-smi
-
       - name: Install dependencies
         run: |
           apt -y update && apt install -y libaio-dev
@@ -270,9 +262,13 @@ jobs:
           rm -rf ~/.cache/torch_extensions/ # shared between conflicting builds
           pip install .[testing,deepspeed,fairscale]
 
-      - name: Are GPUs recognized by our DL frameworks
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
         run: |
-          utils/print_env_pt.py
+          python utils/print_env.py
 
       - name: Run all non-slow selected tests on GPU
         # TODO: Here we pass all tests in the 2 folders for simplicity. It's better to pass only the identified tests.

.github/workflows/self-scheduled.yml

@@ -56,13 +56,6 @@ jobs:
         run: |
           nvidia-smi
 
-      - name: GPU visibility
-        working-directory: /transformers
-        run: |
-          utils/print_env_pt.py
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
-          TF_CPP_MIN_LOG_LEVEL=3 python3 -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
-
   run_tests_single_gpu:
     name: Model tests
     strategy:
@@ -91,6 +84,15 @@ jobs:
         working-directory: /transformers
         run: git fetch && git checkout ${{ github.sha }}
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
       - name: Run all tests on GPU
         working-directory: /transformers
         run: python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }} tests/${{ matrix.folders }}
@@ -135,6 +137,15 @@ jobs:
         working-directory: /transformers
         run: git fetch && git checkout ${{ github.sha }}
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
       - name: Run all tests on GPU
         working-directory: /transformers
         run: python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }} tests/${{ matrix.folders }}
@@ -163,6 +174,15 @@ jobs:
         working-directory: /transformers
         run: git fetch && git checkout ${{ github.sha }}
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
       - name: Run examples tests on GPU
         working-directory: /transformers
         run: |
@@ -197,6 +217,15 @@ jobs:
         working-directory: /transformers
         run: git fetch && git checkout ${{ github.sha }}
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
       - name: Run all pipeline tests on GPU
         working-directory: /transformers
         env:
@@ -233,6 +262,15 @@ jobs:
         run: |
           git fetch && git checkout ${{ github.sha }}
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
       - name: Run all pipeline tests on GPU
         working-directory: /transformers
         env:
@@ -276,6 +314,15 @@ jobs:
           git clone https://github.com/microsoft/DeepSpeed && cd DeepSpeed && rm -rf build
           DS_BUILD_CPU_ADAM=1 DS_BUILD_AIO=1 DS_BUILD_UTILS=1 python3 -m pip install -e . --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
 
+      - name: NVIDIA-SMI
+        run: |
+          nvidia-smi
+
+      - name: Environment
+        working-directory: /workspace/transformers
+        run: |
+          python utils/print_env.py
+
       - name: Run all tests on GPU
         working-directory: /workspace/transformers
         run: |
@@ -293,7 +340,6 @@ jobs:
           name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports
           path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
 
-
   send_results:
     name: Send results to webhook
     runs-on: ubuntu-latest

utils/print_env_pt.py → utils/print_env.py

@@ -17,12 +17,41 @@
 
 # this script dumps information about the environment
 
-import torch
+import os
+import sys
 
+import transformers
 
-print("Torch version:", torch.__version__)
-print("Cuda available:", torch.cuda.is_available())
-print("Cuda version:", torch.version.cuda)
-print("CuDNN version:", torch.backends.cudnn.version())
-print("Number of GPUs available:", torch.cuda.device_count())
-print("NCCL version:", torch.cuda.nccl.version())
+
+os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"
+
+print("Python version:", sys.version)
+print("transformers version:", transformers.__version__)
+
+try:
+    import torch
+
+    print("Torch version:", torch.__version__)
+    print("Cuda available:", torch.cuda.is_available())
+    print("Cuda version:", torch.version.cuda)
+    print("CuDNN version:", torch.backends.cudnn.version())
+    print("Number of GPUs available:", torch.cuda.device_count())
+    print("NCCL version:", torch.cuda.nccl.version())
+except ImportError:
+    print("Torch version:", None)
+
+try:
+    import deepspeed
+
+    print("DeepSpeed version:", deepspeed.__version__)
+except ImportError:
+    print("DeepSpeed version:", None)
+
+try:
+    import tensorflow as tf
+
+    print("TensorFlow version:", tf.__version__)
+    print("TF GPUs available:", bool(tf.config.list_physical_devices("GPU")))
+    print("Number of TF GPUs available:", len(tf.config.list_physical_devices("GPU")))
+except ImportError:
+    print("TensorFlow version:", None)