StudioML A Python model management framework

Simplify and expedite your model building process

What is Studio?

Studio is a model management framework written in Python to help simplify and expedite your model building experience. It was developed to minimize the overhead involved with scheduling, running, monitoring and managing artifacts of your machine learning experiments. No one wants to spend their time configuring different machines, setting up dependencies, or playing archeologist to track down previous model artifacts.

Most of the features are compatible with any Python machine learning framework (Keras, TensorFlow, scikit-learn, etc); some extra features are available for Keras and TensorFlow.

JOIN US ON SLACK!

STAR ON GITHUB

Get Started

Pip install Studio from master PyPi repository:

Run this command:

pip install studioml

For authentication, you must set up an email/password authentication or a Google account. See the docs for more details. For now, get started as a guest by uncommenting "guest:true" in ~/.studio/config.yaml. At this point, you can run a machine learning experiment using Studio. In the example/keras folder, you can find basic examples of experiments to run. Let's try training a Keras model for mnist:

Run this command:

studio run train_mnist_keras.py

Now you can visualize your experiment by navigating to the web interface. Open the web visualizer:

Run this command:

studio ui

You can see the results of your job at http://127.0.0.1:5000. Here you can examine or delete any current or past experiments. Run studio {ui|run} --help for a full list of ui / runner options

Click on your experiment to view experiment artifacts such as the output and workspace, experiment information such as the status and creation time, and the tail of the experiment log.

READ OUR TUTORIAL