https://github.com/automata/mojograd

Implementation of Karpathy's micrograd in Mojo :fire:
https://github.com/automata/mojograd

ai autograd micrograd mojo python

Last synced: about 1 month ago
JSON representation

Implementation of Karpathy's micrograd in Mojo :fire:

• Host: GitHub
• URL: https://github.com/automata/mojograd
• Owner: automata
• Created: 2023-05-29T16:32:25.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-11-19T23:41:54.000Z (about 1 year ago)
• Last Synced: 2024-04-14T20:06:33.536Z (8 months ago)
• Topics: ai, autograd, micrograd, mojo, python
• Language: Python
• Homepage:
• Size: 412 KB
• Stars: 61
• Watchers: 4
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-mojo - mojograd

README

        
# 🔥grad









`mojograd` is a Mojo implementation of [micrograd](https://github.com/karpathy/micrograd),

a reverse-mode autodiff library with a PyTorch-like API.

The goal is to be as close as possible to micrograd, keeping a

pretty clean syntax to define computational graphs.

Like micrograd, it only supports scalar values for now, but we

plan to extend it to support Tensors in the near future.

Note that `mojograd` is in WIP and relies on static register passable structures, so

backward pass copies values and can be really slow (Mojo traits support should

improve that, so please stay tuned!). However, even now with zero

optimizations, forward pass is already ~40x faster than the original Python implementation

(see benchmarks bellow).

# Using

`mojograd` dynamically builds a computational graph by overloading operators 

on `Value` type, performing the forward pass. Just write your expression like

a normal (non-diff) equation and call `backward()` to perform the

backward pass:

```python

from mojograd import Value

var a = Value(2.0)

var b = Value(3.0)

var c: Float32 = 2.0

var d = b**c

var e = a + c

e.backward()

a.print() # => 

b.print() # => 

d.print() # => 

e.print() # =>  

```

For a more complete example (a simple Multi-Layer Perceptron), please check

the `tests.mojo` file. You can run it with:

```bash

mojo tests.mojo

```

# Benchmarks

## MLP binary classifier

When compared to original Python implementation, `mojograd` is up to **~40 times faster

in forward pass**.

| # parameters | micrograd (Python) (sec) | mojograd (Mojo) (sec) | speed up |

|--------------|--------------------------|-----------------------|----------|

| 367          | 0.001                    | 0.00006               | x20      |

| 1185         | 0.004                    | 0.0001                | x40      |

| 4417         | 0.01                     | 0.0005                | x20      |

| 17025        | 0.06                     | 0.002                 | x30      |

# Changelog

- 2023.11.19

  - Benchmarking inference and comparing with micrograd

- 2023.11.18

  - Optimization pass through the code

- 2023.11.14

  - Rebuild the whole thing using pointer handling (dangerous) to register-passables

  - Got the full micrograd implementation working!

  - MLP example training and inference working!

- 2023.09.05

  - Starting from scratch based on suggestions from Jack Clayton

  - Topological sort works but I'm messing something with memory handling,

    the gradients are not getting updated

- 2023.07.04

  - Ported Neuron, Layer and MLP

  - Back to use yakupc55's List (need `register_passable` data struct)

- 2023.06.30

  - Finally got it working! Only missing pow ops and review it