https://github.com/cpldcpu/bitnetpdk

A proof-of-concept (hack) to implement neural network inference on a very tiny 8-bit microcontroller.
https://github.com/cpldcpu/bitnetpdk

free-pdk padauk pfs154 pms150c

Last synced: about 1 year ago
JSON representation

A proof-of-concept (hack) to implement neural network inference on a very tiny 8-bit microcontroller.

• Host: GitHub
• URL: https://github.com/cpldcpu/bitnetpdk
• Owner: cpldcpu
• License: gpl-3.0
• Created: 2024-04-24T20:32:58.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2024-05-05T00:19:52.000Z (about 2 years ago)
• Last Synced: 2025-03-26T10:04:03.297Z (about 1 year ago)
• Topics: free-pdk, padauk, pfs154, pms150c
• Language: C
• Homepage:
• Size: 1.37 MB
• Stars: 68
• Watchers: 5
• Forks: 6
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          

# MNIST inference on the "3 cent" Microcontroller.

This project attempts to push [BitNetMCU](https://github.com/cpldcpu/BitNetMCU) to the absolute limit: Is it possible to implement reasonably accurate inference of MNIST, the handwritten numbers dataset, on a "3 cent" Microcontroller with only 64 bytes of RAM and 1K of instruction memory? 

![banner](docs/banner.png)

Only weeks ago, my resounding answer would have been: "NO". But after the surprisingly good performance of BitNetMCU, I decided to give it a try. 

This code achieves 90.1% test accuracy on test images scaled to 8x8 pixels while consuming 59 of 64 bytes of RAM and 972 of 1024 words of program memory. [See blog](https://cpldcpu.wordpress.com/2024/05/02/machine-learning-mnist-inference-on-the-3-cent-microcontroller/) for a few more details. The training code can be found in [this branch](https://github.com/cpldcpu/BitNetMCU/tree/pdk) of BitNetMCU

## Project structure

```plaintext

├── include/                    # Contains the free-pdk header files

└── src/                

    ├── BitNetMCU_model.h       # MNIST model definition 

    ├── main.c                  # Main source file

    ├── Makefile

    ├── PDK_softuart.c          # Software UART implementation

    └── softuart.h              # Software UART header file

```

    

## Building Instructions

!!! **Use SDCC 4.1.0 to build. Later versions create broken code** !!!

This is a hack, don't expect it to be useful for anything. 

If you still want to build the project, you need to have the [FreePDK toolchain](https://free-pdk.github.io/) installed. The default target is the PFS154, but you can change it to the PMS150C by editing the makefile. Currently, monitoring output is only supported with PFS154 as a target and the UART functions will not be included if PMS150C is selected.

1. Navigate to the `src/` directory.

2. Run `make flash` to build and flash the project

3. `make start` to start the monitor