# SKNet

**Repository Path**: wangchsoft/SKNet

## Basic Information

- **Project Name**: SKNet
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2021-07-19
- **Last Updated**: 2021-07-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# SKNet
Implemenation of [Selective Kernel Networks](https://arxiv.org/abs/1903.06586) by pytorch.

The architecture of **SK** is as follows

![](IMG/SKConv.png)

I trained **SKNet50** on ImageNet-2012 from scratch and got an accuracy of **21.26**, 
which did not reach the performance of **20.79** in the paper.
_If somebody know what caused the problem, please leave me a message._ 

The pretrained weights are provided below.

## Requirement
- `pytorch 1.4.0+`
- `torchvision`
- `tensorboard 1.14+`
- `numpy`
- `pyyaml`
- `tqdm`
- `pillow`


## Dataset
- `ImageNet-2012`


## Pretrained Model on ImageNet-2012

| Architecture | Top-1 error | Pretrained model|
| :----: | :----: | :----: |
| SKNet50 <br> (My Imp.) | 21.26 | [Google Drive](https://drive.google.com/open?id=1h6NIwSemMrFDk4DWT7-Zdm9kolHljyZU)<br>[Baidu Netdisk](https://pan.baidu.com/s/1XTuMDqFuzljxmlfC2TKTyg) |
| SKNet50 <br> ([paper](https://arxiv.org/abs/1903.06586)) | 20.79 | None |


#### If you want to use my pretrained weight, you should do
1. place the downloaded pretrained model in `runs/sknet_imagenet/86028` folder under this project
2. config the attribute of **runid** and **cuda** in the config file `configs/sknet_imagenet.yml`
3. run `validata.py` or `test.py` (For test, you should specify the `img_path` in the [test.py](/test.py))

#### The error curve of SKNet50 during my training process is shown below
<div align="center">
 <img src="IMG/error.png" width = "500" alt="error curve" />
</div>