Skip to content

bk-TurbaAI/codelets_src

 
 

Repository files navigation

Welcome to GeneSys!

The GeneSys compiler uses an embedded architecture description language to create a Hierarchical Architecture Graph for flexibly compiling mg-DFGs to different architectures. The GeneSys architecture uses a systolic array centric approach that forms the core convolution engine for implementing DNN algorithms, and can be customized to run a range of standard DNN topologies.

This document will help you get up and running.

Step 0: Check prerequisites

The following dependencies must be met by your system:

  • python >= 3.7 (For PEP 560 support)

Step 1: Clone the Codelets Src code

$ git clone --recurse-submodules https://github.com/he-actlab/codelets.src
$ cd codelets.src

Step 2: Create a Python virtualenv

Note: You may choose to skip this step if you are doing a system-wide install for multiple users. Please DO NOT skip this step if you are installing for personal use and/or you are a developer.

$ python -m venv general
$ source general/bin/activate
$ python -m pip install pip --upgrade

Step 3: Install GeneSys

If you already have a working installation of Python 3.7 or Python 3.8, the easiest way to install GeneSys is:

$ pip install -e .

Step 4: Compile a benchmark using GeneSys

You can compile a GeneSys benchmark by running the following commands, where <benchmark_name> is one of resnet18, resnet50, or maskrcnn and <output_type> is either "simulation" or "instructions", depending on whether or not the goal is to simulate the compiled output or generate executable instructions:

$ python genesys/benchmarks/run_benchmark.py --benchmark <benchmark_name> --output_type <output_type>

About

Codelet modifed for LLM

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%