upgrade over reinstall (#4667)

* upgrade over reinstall

* update

* mount /var/mlflow_resources to include mlmonitoring

Author: novaturient95

assets/training/automl/environments/ai-ml-automl-dnn-forecasting-gpu/context/Dockerfile

@@ -4,7 +4,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH=/azureml-envs/azureml-automl-dnn-forecasting-
 # Prepend path to AzureML conda environment
 ENV PATH=$AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"

assets/training/automl/environments/ai-ml-automl-dnn-gpu/context/Dockerfile

@@ -6,7 +6,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH=/azureml-envs/azureml-automl-dnn-gpu
 # Prepend path to AzureML conda environment
 ENV PATH=$AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"
@@ -116,7 +116,7 @@ RUN pip install \
                 # end pypi dependencies
 
 # Fix vulnerabilities
-RUN pip install --force-reinstall torchvision==0.23.0 torch==2.8.0 --no-cache-dir
+RUN pip install --upgrade torchvision==0.23.0 torch==2.8.0 --no-cache-dir
 RUN pip install --upgrade 'requests==2.32.5' 'urllib3==2.5.0' 'pillow==12.0.0'
 
 # If you have other python/pip environments, update them too:
@@ -126,9 +126,7 @@ RUN pip install --upgrade 'requests==2.32.5' 'urllib3==2.5.0' 'pillow==12.0.0' |
 # --- End Vulnerability Fixes ---
 
 # --- Pin NumPy <2 everywhere to avoid NumPy 2.x incompatibility ---
-RUN conda run -p $AZUREML_CONDA_ENVIRONMENT_PATH pip install 'numpy==1.23.5' --force-reinstall && \
-    pip install 'numpy==1.23.5' --force-reinstall
-RUN pip install --upgrade 'h2>=4.3.0'
+RUN pip install --upgrade 'numpy==1.23.5' 'h2>=4.3.0'
 
 RUN /bin/bash -c "source activate $AZUREML_CONDA_ENVIRONMENT_PATH && \
  export CUDACXX=/usr/local/cuda/bin/nvcc && \

assets/training/automl/environments/ai-ml-automl-dnn-text-gpu-ptca/context/Dockerfile

@@ -48,18 +48,16 @@ RUN pip install \
 
 RUN pip uninstall -y onnxruntime
 RUN pip uninstall -y torch torchvision torchaudio
-RUN pip install torch==2.8.0  torchaudio==2.8.0 torchvision==0.23.0
-# Install/upgrade torch in ptca env if it exists (fixes vulnerability)
-RUN /opt/conda/envs/ptca/bin/pip install --force-reinstall torch==2.8.0 torchvision==0.23.0 || true
+RUN pip install --upgrade torch==2.8.0  torchaudio==2.8.0 torchvision==0.23.0
 
 # onnx and onnxruntime-training installation
 RUN pip install onnx==1.17.0
 RUN pip uninstall -y onnxruntime-training
 RUN pip install -i https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/onnxruntime-cuda-12/pypi/simple/ onnxruntime-training==1.18.0
 
 # Install numpy
-RUN pip install "numpy==1.23.5" --force-reinstall
-RUN /opt/conda/bin/pip install "numpy==1.23.5" --force-reinstall || true
+RUN pip install --upgrade "numpy==1.23.5"
+RUN /opt/conda/bin/pip install --upgrade "numpy==1.23.5" || true
 
 # torch-ort installation
 RUN TORCH_CUDA_ARCH_LIST="5.2;6.0;7.0;8.0;8.6;9.0" python -m onnxruntime.training.ortmodule.torch_cpp_extensions.install

assets/training/automl/environments/ai-ml-automl-dnn-text-gpu/context/Dockerfile

@@ -4,7 +4,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH=/azureml-envs/azureml-automl-dnn-text-gpu
 # Prepend path to AzureML conda environment
 ENV PATH=$AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"
@@ -129,7 +129,7 @@ RUN pip list && \
                 aiohttp==3.12.14
 
 # Upgrade torch and torchvision
-RUN pip install torch==2.8.0 torchvision==0.23.0 
+RUN pip install --upgrade torch==2.8.0 torchvision==0.23.0 
 RUN /opt/conda/envs/ptca/bin/pip install --upgrade torch==2.8.0 torchvision==0.23.0 || true
 
 # Upgrade starlette and urllib3

assets/training/automl/environments/ai-ml-automl-dnn-vision-gpu/context/Dockerfile

@@ -4,7 +4,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH=/azureml-envs/azureml-automl-dnn-vision-gpu
 # Prepend path to AzureML conda environment
 ENV PATH=$AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"
@@ -130,8 +130,8 @@ RUN /opt/conda/bin/pip install --upgrade 'pillow==12.0.0' 'starlette>=0.49.1' 'u
 RUN /opt/conda/envs/ptca/bin/pip install --upgrade 'pillow==12.0.0' 'starlette>=0.49.1' 'urllib3==2.5.0'
 
 # Upgrade torch and torchvision to fix vulnerabilities
-RUN /opt/conda/envs/ptca/bin/pip install --force-reinstall --no-cache-dir 'torch==2.8.0' torchvision==0.23.0
-RUN pip install --force-reinstall --no-cache-dir 'torch==2.8.0' torchvision==0.23.0
+RUN /opt/conda/envs/ptca/bin/pip install --upgrade --no-cache-dir 'torch==2.8.0' torchvision==0.23.0
+RUN pip install --upgrade --no-cache-dir 'torch==2.8.0' torchvision==0.23.0
 RUN pip install --upgrade 'mlflow-skinny==2.15.1' 'numpy~=1.23.5'
 
 RUN pip install --upgrade 'h2>=4.3.0'

assets/training/automl/environments/ai-ml-automl-dnn/context/Dockerfile

@@ -4,7 +4,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH=/azureml-envs/azureml-automl-dnn
 # Prepend path to AzureML conda environment
 ENV PATH=$AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"

assets/training/automl/environments/ai-ml-automl-gpu/context/Dockerfile

@@ -6,7 +6,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH=/azureml-envs/azureml-automl-dnn-gpu
 # Prepend path to AzureML conda environment
 ENV PATH=$AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"

assets/training/automl/environments/ai-ml-automl/context/Dockerfile

@@ -4,7 +4,7 @@ FROM mcr.microsoft.com/azureml/openmpi5.0-ubuntu24.04:{{latest-image-tag}}
 ENV AZUREML_CONDA_ENVIRONMENT_PATH=/azureml-envs/azureml-automl
 ENV PATH=$AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"

assets/training/forecasting_demand/environments/automl-gpu/context/Dockerfile

@@ -4,7 +4,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH /azureml-envs/automl
 # Prepend path to AzureML conda environment
 ENV PATH $AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"

assets/training/vision/environments/automl-dnn-vision-gpu/context/Dockerfile

@@ -5,7 +5,7 @@ ENV AZUREML_CONDA_ENVIRONMENT_PATH /azureml-envs/azureml-automl-dnn-vision-gpu
 # Prepend path to AzureML conda environment
 ENV PATH $AZUREML_CONDA_ENVIRONMENT_PATH/bin:$PATH
 
-COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/mlflow_score_script.py /var/mlflow_resources/mlflow_score_script.py
+COPY --from=mcr.microsoft.com/azureml/mlflow-ubuntu20.04-py38-cpu-inference:20250506.v1 /var/mlflow_resources/ /var/mlflow_resources/
 
 ENV MLFLOW_MODEL_FOLDER="mlflow-model"
 # ENV AML_APP_ROOT="/var/mlflow_resources"
@@ -78,12 +78,12 @@ RUN conda env create -p $AZUREML_CONDA_ENVIRONMENT_PATH -f conda_dependencies.ya
 
 # vulnearbility fix
 RUN pip install pyarrow==14.0.2
-RUN pip install torch==2.8.0 torchvision==0.23.0
+RUN pip install --upgrade torch==2.8.0 torchvision==0.23.0
 RUN pip install --upgrade urllib3==2.5.0
 
 # Update conda base and ptca envs
 RUN /opt/conda/bin/pip install --upgrade requests urllib3 || true
-RUN /opt/conda/envs/ptca/bin/pip install torch==2.8.0 torchvision==0.23.0 || true
+RUN /opt/conda/envs/ptca/bin/pip install --upgrade torch==2.8.0 torchvision==0.23.0 || true
 RUN /opt/conda/envs/ptca/bin/pip install --upgrade urllib3==2.5.0 || true
 
 # Patch pillow vulnerability