Species Classification Model Training
Workspace
The first step in the training process is to create an Azure Machine Learning workspace. This is done by going to the Azure Portal, and creating a new resource. From here, we can search for Machine Learning, and select Machine Learning. We can then click Create, and fill in the required details. Once the workspace has been created, we can go to the workspace, and click on the Compute tab. From here, we can create a new compute instance, which will be used to run the training script. We can then click on the Compute clusters tab, and create a new compute cluster. This will be used to run the training script. We can then click on the Compute instances tab, and create a new compute instance. This will be used to run the training script.
Our workspace is stored in the following location on Azure cloud storage:
Subscription: natural-state-technical-sub
Resource group: ns-ii-tech-ai-model-rg Workspace name: ns-ii-tech-automl-ws-2
Studio URL: https://ml.azure.com/?tid=bae96e4c-2d85-44f9-a9f8-1fb5efafa959&wsid=/subscriptions/7de38449-8770-4ec9-8a5a-b5b316d96965/resourcegroups/ns-ii-tech-ai-model-rg/providers/Microsoft.MachineLearningServices/workspaces/ns-ii-tech-automl-ws-2
Training Jupyter Notebook
Once the workspace has been created, we can create a new Jupyter Notebook. This is done by going to the workspace, and clicking on the Notebooks tab. From here, we can create a new notebook, and select Python 3.8 - AzureML as the kernel. We can then click Create, and give the notebook a name. We can then click Create, and the notebook will open in a new tab. An example of this notebook can be accessed here: