Skip to content
Programmable Model Compression
Python Jupyter Notebook Shell
Branch: master
Clone or download
Latest commit c7321e0 Nov 6, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
condensa Minor Nov 6, 2019
docs Add GAnalytics tracking Sep 23, 2019
examples/cifar Minor Nov 6, 2019
notebooks Add paper details Nov 7, 2019
test Add CPU tests Oct 16, 2019
.gitignore Update gitignore Nov 6, 2019
.style.yapf Add YAPF config Aug 12, 2019
LICENSE Add License Aug 7, 2019
README.md Add paper details Nov 7, 2019
run_all_tests.sh Add initial tests Aug 15, 2019
setup.cfg Update setup files for PyPi; update README Oct 31, 2019
setup.py Update setup files for PyPi; update README Oct 31, 2019

README.md

A Programming System for Model Compression

Condensa is a framework for programmable model compression in Python. It comes with a set of built-in compression operators which may be used to compose complex compression schemes targeting specific combinations of DNN architecture, hardware platform, and optimization objective. To recover any accuracy lost during compression, Condensa uses a constrained optimization formulation of model compression and employs an Augmented Lagrangian-based algorithm as the optimizer.

Status: Condensa is under active development, and bug reports, pull requests, and other feedback are all highly appreciated. See the contributions section below for more details on how to contribute.

Supported Operators and Schemes

Condensa provides the following set of pre-built compression schemes:

The schemes above are built using one or more compression operators, which may be combined in various ways to define your own custom schemes.

Please refer to the documentation for a detailed description of available operators and schemes.

Prerequisites

Condensa requires:

  • A working Linux installation (we use Ubuntu 18.04)
  • NVIDIA drivers and CUDA 10+ for GPU support
  • Python 3.5 or newer
  • PyTorch 1.0 or newer

Installation

The most straightforward way of installing Condensa is via pip:

pip install condensa

Installation from Source

Retrieve the latest source code from the Condensa repository:

git clone https://github.com/NVlabs/condensa.git

Navigate to the source code directory and run the following:

pip install -e .

Test out the Installation

To check the installation, run the unit test suite:

bash run_all_tests.sh -v

Getting Started

The AlexNet Notebook contains a simple step-by-step walkthrough of compressing a pre-trained model using Condensa. Check out the examples folder for additional, more complex examples of using Condensa (note: some examples require the torchvision package to be installed).

Documentation

Documentation is available here. Please also check out the Condensa paper for a detailed description of Condensa's motivation, features, and performance results.

Contributing

We appreciate all contributions, including bug fixes, new features and documentation, and additional tutorials. You can initiate contributions via Github pull requests. When making code contributions, please follow the PEP 8 Python coding standard and provide unit tests for the new features. Finally, make sure to sign off your commits using the -s flag or adding Signed-off-By: Name<Email> in the commit message.

Citing Condensa

If you use Condensa for research, please consider citing the following paper:

@article{condensa2019,
    title = {A Programmable Approach to Model Compression},
    author = {Joseph, Vinu and Muralidharan, Saurav and Garg, Animesh and Garland, Michael and Gopalakrishnan, Ganesh},
    journal = {CoRR},
    volume = {1911.02497},
    year = {2019},
    url = {https://arxiv.org/abs/1911.02497}
}

Disclaimer

Condensa is a research prototype and not an official NVIDIA product. Many features are still experimental and yet to be properly documented.

You can’t perform that action at this time.