Clone Cerebras Model Zoo#
Cerebras’s Model Zoo repository contains examples of common deep learning models that can be trained on the Cerebras Wafer-Scale Cluster. A list of supported models can be found here.
Use this repository for:
Getting started
Reference implementations in PyTorch of standard deep learning models
Data preparation scripts to use your data
Configuration files for multiple model sizes
Benchmarking
Implementation of standard models optimized for Cerebras hardware
Developing new models
Cerebras-customized PyTorch Layer APIs that simplify the compilation process
Examples that demonstrate best practices for coding your model
Follow the steps on this page to clone the Cerebras Model Zoo repository in your preferred directory.
Clone Cerebras’s Model Zoo#
$ git clone https://github.com/Cerebras/modelzoo.git
Note
If your user node requires a proxy to access to external packages, you will need to configure a proxy in your git configuration to clone any repository.
Set up environment for Cerebras Model Zoo#
Once you have Set up Cerebras’s virtual environment, install additional packages to your Python environment to support all the functionalities of the Cerebras Model Zoo.
Install packages available in Cerebras Model Zoo by issuing the following commands:
$ source venv_cerebras_pt/bin/activate
(venv_cerebras_pt) $ cd modelzoo
(venv_cerebras_pt) $ pip install -r requirements_cb_pt.txt
Note
Refer to Run Cerebras Model Zoo on a GPU if you are running your model on a GPU.