chevron icon
All
Machine Learning

Create an API endpoint for a model

Philip Kiely
April 22, 2022

When you deploy a model on Baseten, you can invoke it via an API right away. This saves you the effort of building and deploying a model server in Flask or Django, making it faster to interact with models. But in production use cases, you often need to wrap model invocation in parsing and business logic for the model to be useful. In general, ML models enjoy big blobs of JSON; humans enjoy meaningful sentences. With worklets, you can build API endpoints with custom code around model invocations.

In this tutorial, we’ll be building an API endpoint for the English to French translation model from our pre-trained model library. All you’ll need is an api key. You can call the API endpoints from the command line, your microservice of choice, or Postman.

Model-based endpoint

On the models page, click “Deploy a model” then select “Use a pre-trained model.” Scroll to “English to French translation, select it, click “Next,” then “Deploy model.” From the deployed model page, you can see a command calling the model via curl.

Here is a model input:

{
    "inputs": [
        "She put the cat in her purse somebody come get her!!",
        "I just want someone who looks at me the way my brother looks at ketchup",
        "my goal for our friendship is to reach a point where im fb friends with your mom"
    ]
}

Which returns:

{
    "model_id": "VqmWa0O",
    "model_version_id": "7qrjz13",
    "predictions":
    [
        {
            "translation_text": "Elle a mis le chat dans sa poche quelqu'un vient la chercher!!"
        },
        {
            "translation_text": "Je veux simplement quelqu'un qui me regarde comme mon frère regarde le ketchup"
        },
        {
            "translation_text": "mon objectif pour notre amitié est d'atteindre un point où j'ai des amis fb avec votre mère"
        }
    ]
}

Worklet-based endpoint

The previous API endpoint returns a valid translation, but you have to match up the translations to the inputs yourself. With a worklet, we can implement cleaner inputs and outputs.

In the starter application that we deployed earlier, visit the “Translate sentences” worklet. If you're curious, the code driving these changes is in main.py. Click on “API Endpoint” to see what to copy to call the worklet endpoint.

Here is a worklet input: 

{
    "worklet_input":
    {
        "sentences":
        [
            "She put the cat in her purse somebody come get her!!",
            "I just want someone who looks at me the way my brother looks at ketchup",
            "my goal for our friendship is to reach a point where im fb friends with your mom"
        ]
    }
}

 Which returns:

{
    "success": true,
    "worklet_output":
    [
        {
            "original": "She put the cat in her purse somebody come get her!!",
            "translation": "Elle a mis le chat dans sa poche quelqu'un vient la chercher!!"
        },
        {
            "original": "I just want someone who looks at me the way my brother looks at ketchup",
            "translation": "Je veux simplement quelqu'un qui me regarde comme mon frère regarde le ketchup"
        },
        {
            "original": "my goal for our friendship is to reach a point where im fb friends with your mom",
            "translation": "mon objectif pour notre amitié est d'atteindre un point où j'ai des amis fb avec votre mère"
        }
    ]
}
}

Explore more

Product

Fine-tune FLAN-T5 on Blueprint today!

You can now fine-tune FLAN-T5, an instruction-tuned text-to-text transformer model developed by Google, on Blueprint!

Abu Qader
March 17, 2023
Product

Baseten achieves SOC 2 Type II certification

Baseten has completed a rigorous audit by Sensiba San Filippo LLP, a leading CPA firm, to achieve System and Organization Controls (SOC) 2 Type II certification.

Baseten
March 8, 2023
SOC 2 Type II