Welcome to MateVerse API documentation

The Mateverse API empowers you to do the Machine Learning and Deep Learning predictions using few simple HTTP requests. The API offers image and text classification (as of now) as a service. Using our API you are only one step away from the awesome world of ML/AI

The API usage is as simple as it can get. You just need to input the data that you have into your model and you will get the predictions out of it.

Here is a sneak peak into the API details: models

Prediction

The Predict API call will return you the confidence score of your images that are trained on a particular model of your choice. When you make a prediction through the API, you pass in the image and tell it what model to use. There are several ways of using the Prediction API

Get Prediction via uploading through local system

To use any model and get the prediction from it, you just need to supply an image and specify the model that you would like to use. You can send as many number of images in one API call. You can specify the api_secret key, the model_id and the path to the file and you are ready to make a API call.

Consider the dummy example to get a thorough understanding of it:

curl -F "api_secret=xxxxxxxxxxxx" \
-F "model_id=xxx" \
-F "file=@/path/images.jpeg" https://www.mateverse.com/v1/predict/

API Response

Following is the response you get when you run the Prediction API

[
  {
    "images.jpg": [
      {
        "predicted_score": "0.95155",
        "predicted_label": "bracelet"
      },
      {
        "predicted_score": "0.04845",
        "predicted_label": "earing"
      }
    ]
  }
]

The predicted_score is the confidence score of the image and the predicted_label is the classified class. So according to the above API call, 95.15% chances are that the given image is a bracelet. Hola, indeed it is.

So here we are, making a single line of API call to use Image_Classification model. The power of Machine Learning in a single line of code.

Go ahead and predict your model now.