Data Asset Registration

# Create data asset using Azure CLI
az ml data create \
  --name retail_data \
  --version 1 \
  --type uri_folder \
  --path "azureml://datastores/workspaceblobstore/paths/raw/ecommerce/" \
  --description "Retail e-commerce dataset 2019 - Raw CSV files for training" \
  --tags Environment=development Dataset=ecommerce Year=2019

# List data assets
az ml data list --name retail_data --output table

# Show specific asset details
az ml data show --name retail_data --version 1

# Verify asset creation
az ml data show --name retail_data --version 1 --query "{name:name, version:version, type:type, path:path}"

# Check asset size and file count
az ml data show --name retail_data --version 1 --query "{size:size, fileCount:fileCount}"

# Python script to test data asset accessibility
from azure.ai.ml import MLClient
from azure.ai.ml.entities import Data
from azure.identity import DefaultAzureCredential
import os

def test_data_asset():
    # Initialize ML client
    credential = DefaultAzureCredential()
    ml_client = MLClient(
        credential=credential,
        subscription_id="your-subscription-id",
        resource_group_name="retail-dev-rg",
        workspace_name="retail-ml-workspace"
    )
    
    # Get data asset
    data_asset = ml_client.data.get(name="retail_data", version="1")
    
    print(f"📊 Data Asset Information:")
    print(f"  Name: {data_asset.name}")
    print(f"  Version: {data_asset.version}")
    print(f"  Type: {data_asset.type}")
    print(f"  Path: {data_asset.path}")
    print(f"  Size: {data_asset.size} bytes")
    
    # Test data access
    try:
        # List files in the asset
        files = ml_client.data.list(name="retail_data", version="1")
        print(f"\n📁 Files in dataset:")
        for file in files:
            print(f"  ✅ {file.name}")
        
        print(f"\n✅ Data asset is accessible and ready for training jobs!")
        
    except Exception as e:
        print(f"❌ Error accessing data asset: {e}")

if __name__ == "__main__":
    test_data_asset()

# Test data asset in training job context
from azure.ai.ml import MLClient, command
from azure.ai.ml.entities import Data
from azure.identity import DefaultAzureCredential

def create_test_job():
    credential = DefaultAzureCredential()
    ml_client = MLClient(
        credential=credential,
        subscription_id="your-subscription-id",
        resource_group_name="retail-dev-rg",
        workspace_name="retail-ml-workspace"
    )
    
    # Create test training job using the data asset
    job = command(
        code="./src",
        command="python train.py --data ${{inputs.retail_data}}",
        inputs={
            "retail_data": Data(
                type="uri_folder",
                path="azureml://datastores/workspaceblobstore/paths/raw/ecommerce/"
            )
        },
        environment="azureml://registries/azureml/environments/sklearn-1.0/labels/latest",
        compute="cpu-cluster",
        display_name="test-retail-data-access"
    )
    
    print("✅ Training job created successfully with data asset reference!")
    return job

if __name__ == "__main__":
    create_test_job()

#!/bin/bash
# create-data-asset.sh

# Configuration
SUBSCRIPTION_ID="your-subscription-id"
RESOURCE_GROUP="retail-dev-rg"
WORKSPACE_NAME="retail-ml-workspace"
ASSET_NAME="retail_data"
ASSET_VERSION="1"
STORAGE_ACCOUNT="retailforecastdev"
CONTAINER="data"
FOLDER_PATH="raw/ecommerce"

echo "🚀 Creating Data Asset in Azure ML Studio..."

# Login to Azure
az login

# Set subscription
az account set --subscription $SUBSCRIPTION_ID

# Create data asset
az ml data create \
  --name $ASSET_NAME \
  --version $ASSET_VERSION \
  --type uri_folder \
  --path "azureml://datastores/workspaceblobstore/paths/$FOLDER_PATH/" \
  --description "Retail e-commerce dataset 2019 - Raw CSV files for training" \
  --tags Environment=development Dataset=ecommerce Year=2019

# Verify creation
echo "✅ Verifying data asset creation..."
az ml data show --name $ASSET_NAME --version $ASSET_VERSION --output table

echo "📊 Data Asset Information:"
az ml data show --name $ASSET_NAME --version $ASSET_VERSION --query "{name:name, version:version, type:type, path:path, size:size}"

echo "✅ Data asset created successfully!"

# create_data_asset.py
from azure.ai.ml import MLClient
from azure.ai.ml.entities import Data
from azure.identity import DefaultAzureCredential
import os

def create_data_asset():
    # Configuration
    subscription_id = "your-subscription-id"
    resource_group = "retail-dev-rg"
    workspace_name = "retail-ml-workspace"
    
    # Initialize ML client
    credential = DefaultAzureCredential()
    ml_client = MLClient(
        credential=credential,
        subscription_id=subscription_id,
        resource_group_name=resource_group,
        workspace_name=workspace_name
    )
    
    # Create data asset
    data_asset = Data(
        name="retail_data",
        version="1",
        type="uri_folder",
        path="azureml://datastores/workspaceblobstore/paths/raw/ecommerce/",
        description="Retail e-commerce dataset 2019 - Raw CSV files for training",
        tags={
            "Environment": "development",
            "Dataset": "ecommerce",
            "Year": "2019"
        }
    )
    
    # Register the asset
    ml_client.data.create_or_update(data_asset)
    
    print("✅ Data asset created successfully!")
    
    # Verify creation
    created_asset = ml_client.data.get(name="retail_data", version="1")
    print(f"📊 Asset Details:")
    print(f"  Name: {created_asset.name}")
    print(f"  Version: {created_asset.version}")
    print(f"  Type: {created_asset.type}")
    print(f"  Path: {created_asset.path}")

if __name__ == "__main__":
    create_data_asset()

# Verify asset exists
az ml data list --name retail_data

# Check workspace context
az ml workspace show --name retail-ml-workspace --resource-group retail-dev-rg

# Check Managed Identity permissions
az role assignment list --assignee <ml-workspace-mi-id> --scope /subscriptions/<sub-id>/resourceGroups/retail-dev-rg/providers/Microsoft.Storage/storageAccounts/retailforecastdev

# Verify storage account and container
az storage container show --name data --account-name retailforecastdev

# Check folder path exists
az storage blob list --container-name data --prefix "raw/ecommerce/" --account-name retailforecastdev