Follow the steps below to get started with the free version of PerceptiLabs. Contact us for information about installing our Docker and OpenShift versions.
Step 1: Install and Run PerceptiLabs
1. Install a version of Python that meets our Python Requirements. Since multiple versions of Python and PerceptiLabs can be installed concurrently, we recommend you install them in a virtual Python environment as described here.
2. Run one of the following:
pip install perceptilabs
Alternatively you can install our latest nightly build of the CPU-only version with pip install pl-nightly. Note that the nightly build may not be as stable as the latest official release.
pip install perceptilabs-gpu
Check out this forum topic for information on setting up the GPU version in Anaconda to simplify installation.
3. Run perceptilabs. The app will open in your default web browser.
If you're using Chrome, first ensure that hardware acceleration is enabled to ensure the best performance of PerceptiLabs on this browser.
To enable this, click on Chrome's triple-dot menu and select Settings. Then navigate to Advanced > System and enable Use hardware acceleration when available; you may need to relaunch your browser as well.
Step 2: Sign in
Within PerceptiLabs, sign up for a free PerceptiLabs account and then sign in to start building your machine learning models with PerceptiLabs!
The Kernel must be running in order to access PerceptiLabs and its sign-up screen
Step 3: Watch this video
Watch the following short video that shows how to create your first model from scratch: