Skip to main content
Join the official 2019 Python Developers SurveyStart the survey!

Blazee makes it easy to deploy Machine Learning models on the cloud and turn them into an awesome prediction API.

Project description

Python Library for Blazee

Intro

Blazee is the easiest and fastest way to turn your Machine Learning models and pipelines into a production ready prediction API.

It allows you to deploy trained models straight from a Jupyter Notebook or any other model training environment, and access them live from anywhere using the Blazee HTTP API.

This library can also be used

Supported Frameworks

At the moment, we support the following frameworks:

  • Scikit Learn (Supervised learning models and pipeline)
  • Keras
  • PyTorch
  • XGBoost
  • LightGBM

Coming soon:

  • H2O
  • XGBoost
  • LightGBM
  • Tensorflow

Installation

Install from pip

pip install blazee

Sign up and get an API Key from https://blazee.io

Usage

# Train your model like you usually do
>>> from sklearn.linear_model import LogisticRegressionCV
>>> clf = LogisticRegressionCV()
>>> ...
>>> clf.train(X)

# Deploy your model on Blazee
# Get your API Key on https://blazee.io
>>> from blazee import Blazee
>>> bz = Blazee(YOUR_API_KEY)
>>> model = bz.deploy_model(clf)
Uploading model to Blazee...
Successfully deployed model bdea76f4-fa0f-4ef1-8bc5-f36978a4488e
Deploying model... This will take a few moments

# Predict a single sample
>>> pred = model.predict(X[0])
>>> pred.prediction
1
>>> pred.probas
{0: 0.08, 1:0.91, 2: 0.01}

# Or predict a batch
>>> preds = model.batch_predict(X)

# Deploy another version of the model
>>> clf2 = SGDClassifier()
>>> ...
>>> clf2.train(X)
>>> model.update(clf2)

Support

Contact us at support@blazee.io or open a Github Issue for any question or bug report.

Project details


Download files

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

Files for blazee, version 0.1.5
Filename, size File type Python version Upload date Hashes
Filename, size blazee-0.1.5-py3-none-any.whl (28.3 kB) File type Wheel Python version py3 Upload date Hashes View hashes
Filename, size blazee-0.1.5.tar.gz (12.4 kB) File type Source Python version None Upload date Hashes View hashes

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page