SODA-BENCHMARKS

This project contains scripts and code to benchmark SODA tools in different scenarios.

How to Use?

We depend on Docker and a pre-built Docker image which contains the binaries to run the benchmarks. We recommend using VS Code Dev Containers to orchestrate the container setup and provide the correct environment for running the examples and benchmarks.

To use Dev Containers, follow these steps:

1. Install Visual Studio Code: If you haven't already, download and install Visual Studio Code from here.
2. Install the Container Tools extension: Open Visual Studio Code and install the "Container Tools" extension from the Extensions view.
3. Open the folder in a Dev Container: Press Ctrl+Shift+P (or Cmd+Shift+P on macOS) and select: Dev Containers: Reopen in Container.
4. Navigate to the relevant folder and check its README.md for instructions on executing the examples or benchmarks.

What Next?

Start by navigating to the examples folder. Each example includes a README.md with instructions on how to run it and details about the generated artifacts.

For a step-by-step guide, see the tutorials folder. It includes Jupyter notebooks that demonstrate the full workflow, from PyTorch models to Verilog and GDS generation using SODA and open-source tools.

If you want to create accelerator cores for your own models, check out the models folder. It contains Python scripts to download or implement machine learning models, and scripts to transform these models or parts of them into Verilog.

Project Structure

├── docs         # Documentation
├── examples     # Simple examples demonstrating the end-to-end Python to Verilog/GDS flow
├── LICENSE
├── models       # Python scripts to download or implement ML models, and scripts to transform models into Verilog
├── README.md
├── scripts      # Bash and Makefile scripts used in other folders
├── tests        # All tests (if possible)
├── tutorials    # Guided tutorials using jupyter notebooks
|
├── external     # Contains source code from third party projects
└── builds       # Gets created with scritps/external.setup-<projname>.sh scripts

License

This project is made available under the Apache License 2.0 with LLVM Exceptions. See the LICENSE file for more details.

Software from third parties included in the soda-benchmarks project

The soda-benchmarks project contains third party software which is under different license terms. All such code will be identified clearly using at least one of two mechanisms:

1. It will be in a separate directory tree with its own LICENSE.txt or LICENSE file at the top containing the specific license and restrictions which apply to that software, or
2. It will contain specific license and restriction terms at the top of every file.

Disclaimer Notice

This material was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor the United States Department of Energy, nor Battelle, nor any of their employees, nor any jurisdiction or organization that has cooperated in the development of these materials, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness or any information, apparatus, product, software, or process disclosed, or represents that its use would not infringe privately owned rights.

Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or Battelle Memorial Institute. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

PACIFIC NORTHWEST NATIONAL LABORATORY
operated by
BATTELLE
for the
UNITED STATES DEPARTMENT OF ENERGY
under Contract DE-AC05-76RL01830