Frank Odom (@fkodom)
github

fkodom

Twitter logo Share LinkedIn logo Share Facebook logo Share
  • Stars
    star
    1,120
  • Global Rank 27,450 (Top 1.0 %)
  • Followers 287
  • Following 29
  • Registered almost 6 years ago
  • Most used languages
    Python
         95.8 %
    Dockerfile
         4.2 %
  • Location 🇺🇸 United States
  • Country Total Rank 12,946
  • Country Ranking
    Python
         1,346
    Dockerfile
         4,632

Top repositories

1

fft-conv-pytorch

Implementation of 1D, 2D, and 3D FFT convolutions in PyTorch. Much faster than direct convolutions for large kernel sizes.
Python
472
star
2

grouped-query-attention-pytorch

(Unofficial) PyTorch implementation of grouped-query attention (GQA) from "GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints" (https://arxiv.org/pdf/2305.13245.pdf)
Python
118
star
3

yet-another-retnet

A simple but robust PyTorch implementation of RetNet from "Retentive Network: A Successor to Transformer for Large Language Models" (https://arxiv.org/pdf/2307.08621.pdf)
Python
101
star
4

clip-text-decoder

Generate text captions for images from their embeddings.
Python
97
star
5

transformer-from-scratch

Code implementation from my blog post: https://fkodom.substack.com/p/transformers-from-scratch-in-pytorch
Python
90
star
6

soft-mixture-of-experts

PyTorch implementation of Soft MoE by Google Brain in "From Sparse to Soft Mixtures of Experts" (https://arxiv.org/pdf/2308.00951.pdf)
Python
64
star
7

dilated-attention-pytorch

(Unofficial) Implementation of dilated attention from "LongNet: Scaling Transformers to 1,000,000,000 Tokens" (https://arxiv.org/abs/2307.02486)
Python
49
star
8

wnet-unsupervised-image-segmentation

Python
26
star
9

lora-pytorch

Simple but robust implementation of LoRA for PyTorch. Compatible with NLP, CV, and other model types. Strongly typed and tested.
Python
19
star
10

semantle

Fastest 'Semantle' solver this side of the Mississippi.
Python
15
star
11

python-repo-template

Template repo for Python projects, especially those focusing on machine learning and/or deep learning.
Python
12
star
12

document-rag

RAG application to answer questions about PDF documents using LLMs.
Python
11
star
13

byol

PyTorch implementation of BYOL: a fantastically simple method for self-supervised image representation learning with SOTA performance.
Python
9
star
14

simple-bert-pytorch

A very simple BERT implementation in PyTorch, which only depends on PyTorch itself.
Python
6
star
15

blip-inference

Pretrained BLIP with a similar API to CLIP.
Python
5
star
16

octconv-pytorch

Python
5
star
17

metis

Minimalist library for training RL agents in PyTorch. Implements many common training algorithms, with a focus on actor-critic methods. Includes SAC, TD3, PPO, A2C, VPG.
Python
5
star
18

simple-cocotools

A simple, modern alternative to 'pycocotools'.
Python
3
star
19

unipipe

Build batch pipelines in Python that run anywhere.
Python
3
star
20

ncsn

Personal repo recreating Noise Contrastive Score Networks (https://arxiv.org/abs/1907.05600v2) for learning purposes.
Python
2
star
21

wordle

Fastest Wordle solver in the West.
Python
2
star
22

docker-images

Repository for automatically building base Docker images with Github Actions.
Dockerfile
1
star
23

flaxseed

(Personal) Training library build on top of Flax, making it easier to train deep learning models with Jax. Because Flax is great, but doesn't take things far enough.
Python
1
star
24

ddpg-her-pytorch

Implementation of the Hindsight Experience Replay (HER) algorithm using PyTorch. Utilizes Deep Deterministic Policy Gradients for off-policy optimization of RL agents in continuous action spaces.
Python
1
star