Azure AI & ML Studio

from openai import AzureOpenAI

client = AzureOpenAI(
    azure_endpoint="https://YOUR-RESOURCE.openai.azure.com/",
    api_key="YOUR-KEY",
    api_version="2024-02-01"
)

response = client.chat.completions.create(
    model="gpt-4o",          # Your deployment name
    messages=[
        {"role": "system", "content": "You are a helpful AI assistant."},
        {"role": "user", "content": "Explain transformer architecture."}
    ],
    max_tokens=512
)
print(response.choices[0].message.content)

from azure.ai.ml import MLClient, command
from azure.ai.ml.entities import Environment, AmlCompute
from azure.identity import DefaultAzureCredential

ml_client = MLClient(DefaultAzureCredential(), subscription_id, rg, workspace)

# Define training job
job = command(
    code="./src",                    # Local code directory
    command="python train.py --epochs 50",
    environment="AzureML-pytorch-1.13-ubuntu20.04-py38-cuda11-gpu@latest",
    compute="gpu-cluster",           # Your compute cluster name
    experiment_name="my-experiment",
    display_name="training-run-1"
)

returned_job = ml_client.jobs.create_or_update(job)
ml_client.jobs.stream(returned_job.name)  # Stream logs

import mlflow

# Azure ML auto-configures MLflow tracking — just log!
with mlflow.start_run():
    mlflow.log_param("learning_rate", 0.001)
    mlflow.log_param("epochs", 50)

    for epoch in range(50):
        train_loss = train_one_epoch()
        val_accuracy = evaluate()

        mlflow.log_metric("train_loss", train_loss, step=epoch)
        mlflow.log_metric("val_accuracy", val_accuracy, step=epoch)

    # Save model
    mlflow.pytorch.log_model(model, "model")