Skip to content

AzureML

Adrian Boguszewski edited this page Apr 26, 2022 · 37 revisions

Instructions to Run the Notebooks in Azure ML Studio

The steps below assume that you have an Azure account and access to the Azure ML Studio. The entire one-time setup process may take up to 20 minutes.

Step 0: Add a Compute Instance

In Azure ML Studio, add a compute instance and pick any CPU-based instance (No GPU required, but we recommend at least 4 CPU cores and 8GB of RAM).
azure1

Once the compute instance is running, open the terminal and then run Steps 1-8 below. azure2a

Step 1: Create a Conda Environment

conda create --name openvino_env python=3.8 -y

Step 2: Activate the Environment

conda activate openvino_env

Step 3: Clone OpenVINO Notebooks

git clone https://github.com/openvinotoolkit/openvino_notebooks.git

Step 4: Change Directory to openvino_notebooks

cd openvino_notebooks

Step 5: Upgrade pip and Install Requirements

python -m pip install --upgrade pip
pip install -r requirements.txt

Step 6: Add openvino_env to PATH

set PATH="/anaconda/envs/openvino_env/bin;%PATH%"

Step 7: Run the Notebooks!

To run the notebooks, click on Notebooks and refresh your Files:
azure3a

Select a notebook:
azure4

Then run all cells:
azure6a

Note: Please also make sure you are using the 'openvino_env' environment (not Python 3). That's all :) Happy coding. image