TransCNN-HAE

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #typescript #llm #go #golang #android #machine-learning #rust #deep-learning #linux

(★ 13)

No description available.

TransCNN-HAE 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# G2LFormer: Global-to-Local Token Mixing Transformer for Blind Image Inpainting and Beyond
Haoru Zhao*, Zonghui Guo*, Shishi Qiao, Zhaorui Gu, Bing Zheng, Junyu Dong, Haiyong Zheng<br>

Here we provide the PyTorch implementation of our latest version, if you require the code of our previous ACM MM version (**["TransCNN-HAE: Transformer-CNN Hybrid AutoEncoder for Blind Image Inpainting"](https://dl.acm.org/doi/pdf/10.1145/3503161.3547848)**), please click the **[released version](https://github.com/zhenglab/TransCNN-HAE/releases/tag/v1.0)**.

## Prerequisites

- Linux
- Python 3.7
- NVIDIA GPU + CUDA CuDNN

## Getting Started

### Installation

- Clone this repo:
```bash
git clone https://github.com/zhenglab/TransCNN-HAE.git
cd TransCNN-HAE
```

- Install [PyTorch](http://pytorch.org) and 1.7 and other dependencies (e.g., torchvision).
  - For Conda users, you can create a new Conda environment using `conda create --name <env> --file requirements.txt`.

### Training

- Train our TransCNN-HAE+:

```
python train.py --path=./checkpoints/config.yml
```

- Train our TransCNN-HAE+wCIA:

```
python train.py --path=./checkpoints/configwithCIA.yml
```

### Testing

The model is automatically saved every 10,000 iterations, please rename the file `g.pth_$iter_number$` to `g.pth` and then run testing command.

- Test our TransCNN-HAE+:

```
python test.py --path=./checkpoints/config.yml
```

- Test our TransCNN-HAE+wCIA:

```
python test.py --path=./checkpoints/configwithCIA.yml
```

### Pre-trained Models

- Download pre-trained models from [BaiduCloud](https://pan.baidu.com/s/1PrHhJ-VSPV9v-ojJr0Lm_A?pwd=3w72) (access code: 3w72), and put `g.pth` in the directory `checkpoints`.

TransCNN-HAE

TransCNN-HAE 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

TransCNN-HAE

TransCNN-HAE 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README