Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[NeurIPS 2022] HorNet: Efficient High-Order Spatial Interactions with Recursive Gated Convolutions
| Date | Stars |
|---|---|
| 2026-07-24 | 345 |
| 2026-07-25 | 345 |
| 2026-07-28 | 345 |
| 2026-07-30 | 345 |
| 2026-08-06 | 345 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# HorNet <img width="32" alt="HorNet Icon" src="figs/hornet_icon.png"> Created by [Yongming Rao](https://raoyongming.github.io/)\*, [Wenliang Zhao](https://wl-zhao.github.io/)\*, [Yansong Tang](https://andytang15.github.io/), [Jie Zhou](https://scholar.google.com/citations?user=6a79aPwAAAAJ&hl=en&authuser=1), [Ser-Nam Lim](https://sites.google.com/site/sernam)†, [Jiwen Lu](https://scholar.google.com/citations?user=TN8uDQoAAAAJ&hl=en&authuser=1)† This repository contains PyTorch implementation for HorNet (NeurIPS 2022). HorNet is a family of generic vision backbones that perform explicit *high-order* spatial interactions based on Recursive Gated Convolution.  [[Project Page]](https://hornet.ivg-research.xyz/) [[arXiv]](https://arxiv.org/abs/2207.14284) ## Model Zoo ImageNet-1K trained models: | name | arch | Params | FLOPs | Top-1 | url | | --- | --- | --- | --- | --- | --- | | HorNet-T (7x7) | ```hornet_tiny_7x7``` | 22M | 4.0G | 82.8 | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/bac9690c72944d19aca1/?dl=1)| | HorNet-T (GF) | ```hornet_tiny_gf``` | 23M | 3.9G | 83.0 | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/460e759c82324ea6bfa3/?dl=1)| | HorNet-S (7x7) | ```hornet_small_7x7``` | 50M | 8.8G | 83.8 | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/8ef64277da0e469ea857/?dl=1)| | HorNet-S (GF) | ```hornet_small_gf``` | 50M | 8.7G | 84.0 | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/959603ca287d4c09ad49/?dl=1)| | HorNet-B (7x7) | ```hornet_base_7x7``` | 87M | 15.6G | 84.2 | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/13a748487711492e9704/?dl=1)| | HorNet-B (GF) | ```hornet_base_gf``` | 88M | 15.5G | 84.3 | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/c212d1d8a3914b47a5c1/?dl=1)| ImageNet-22K trained models: | name | arch | Params | FLOPs | url | | --- | --- | --- | --- | --- | | HorNet-L (7x7) | ```hornet_large_7x7``` | 209M | 34.8G | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/c2fc23caae084a3aa73d/?dl=1)| | HorNet-L (GF) | ```hornet_large_gf``` | 211M | 34.7G | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/8600206ebdfa485f98d0/?dl=1)| | HorNet-L (GF)* | ```hornet_large_gf_img384``` | 216M | 101.8G | [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/eec48d5eca5a48d98fd2/?dl=1)| *indicate the model is finetuned to 384x384 resolution on ImageNet-22k. ## ImageNet Classification ### Requirements - torch==1.8.0 - torchvision==0.9.0 - timm==0.4.12 - tensorboardX - six - submitit (multi-node training) **Data preparation**: download and extract ImageNet images from http://image-net.org/. The directory structure should be ``` │ILSVRC2012/ ├──train/ │ ├── n01440764 │ │ ├── n01440764_10026.JPEG │ │ ├── n01440764_10027.JPEG │ │ ├── ...... │ ├── ...... ├──val/ │ ├── n01440764 │ │ ├── ILSVRC2012_val_00000293.JPEG │ │ ├── ILSVRC2012_val_00002138.JPEG │ │ ├── ...... │ ├── ...... ``` ### Evaluation To evaluate a pre-trained HorNet model on the ImageNet validation set with 8 GPUs, run: ``` python -m torch.distributed.launch --nproc_per_node=8 main.py \ --model hornet_tiny_7x7 --eval true --input_size 224 \ --resume /path/to/checkpoint \ --data_path /path/to/imagenet-1k ``` ### Training To train HorNet models on ImageNet from scratch on a single machine, run: ``` python -m torch.distributed.launch --nproc_per_node=8 main.py \ --model hornet_tiny_7x7 --drop_path 0.2 --clip_grad 5\ --batch_size 128 --lr 4e-3 --update_freq 4 \ --model_ema true --model_ema_eval true \ --data_path /path/to/imagenet-1k \ --output_dir ./logs/hornet_tiny_7x7 ``` We provide detailed training commands for our models in [TRAINING.md](./TRAINING.md). ## Downstream Tasks Please check the [object_detection.md](./object_detection/README.md) and [semantic_segmentation.md](./semantic_segmentation/README.md) for training and evaluation instructions on dense prediction tasks. HorNet also achieves state-of-the-art performance on 3D object classification with our new framework ([
Excerpt of 5,232 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:790b5e0c0f053930, topic:computer-vision, readme:object detection, readme:semantic segmentation
matched fp:790b5e0c0f053930, topic:pytorch