Making Predictions

Discover predictors and make predictions.

Function provides a Python client for making predictions in any Python app.

Installing Function

To start, open a terminal window and run the following command:

# Open a terminal and run the following command:
pip install --upgrade fxn

Logging in to Function

First, head over to your Account page to generate an access key:

generate access key

Now, let's login to Function with your access key:

from fxn import Function
# Create the Function client with your access key
# Or set the `FXN_ACCESS_KEY` environment variable
fxn = Function(access_key="<ACCESS KEY>")

If you don't provide an access_key to the Function client, the FXN_ACCESS_KEY environment variable will be used.

Making Predictions

First, we'll need a predictor to make predictions with. You can either use a public predictor on Function, or make your own. In this guide, we'll be using the @samples/greeting predictor:

You can discover and use public predictors on Function.

We'll be using the greeting predictor in this guide.

The predictor accepts a name of the person to greet, optionally with their age and city. Let's make the prediction:

# Predict
prediction = fxn.predictions.create(
tag="@samples/greeting",
inputs={
"name": "Rhea",
"age": 44,
"city": "Los Angeles"
}
)
# Print
print(prediction.results[0])

Function supports serializing audio, images, video, tensors, and much more. For more info, see Value

Previous
Core Concepts