Skip to main content

Create controlled and compliant AI systems with PredictionGuard.

Project description

Prediction Guard - Python Client

This package provides functionality developed to simplify interfacing with Prediction Guard in Python 3.


See the API documentation.


The package can be installed with pip:

pip install --upgrade predictionguard


  • Python 3.6+

Quick Start

To use this library, you must have an access token and specify it as a string when creating the pg.Client object. Access tokens can be created through the Prediction Guard platform (link coming soon). This is a basic example that:

  1. Instantiates a Prediction Guard client
  2. Defines some example model input/ output
  3. Creates a prediction proxy endpoint
  4. Uses the endpoint to make a prediction
import predictionguard as pg

# Initialize a Prediction Guard client.
client = Client(token=<your access token>)

# Create some examples illustrating the kind of predictions you
# want to make (domain/ use case specific).
examples = [
   "input": {
     "phrase": "I'm so excited about Prediction Guard. It's gonna be awesome!"
   "output": {
     "sentiment": "POS"
   "input": {
     "phrase": "AI development without Prediction Guard is bad. It's really terrible."
   "output": {
     "sentiment": "NEG"

# Create a prediction "proxy." This proxy will save your examples, evaluate
# SOTA models to find the best one for your use case, and expose the best model
# at an endpoint corresponding to the proxy.
client.create_proxy(task='sentiment', name='my-sentiment-proxy', examples=examples)

# Now your ready to start getting reliable, future proof predictions. No fuss!
client.predict(name='test-client-sentiment4', data={
 "phrase": "Isn't this great! I'm so happy I'm using Prediction Guard"

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

predictionguard-1.9.4.tar.gz (6.8 kB view hashes)

Uploaded Source

Built Distribution

predictionguard-1.9.4-py2.py3-none-any.whl (6.4 kB view hashes)

Uploaded Python 2 Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page