Skip to content

Latest commit

 

History

History
54 lines (41 loc) · 1.76 KB

INSTALL.md

File metadata and controls

54 lines (41 loc) · 1.76 KB
Raw

Installation

Dependencies Installation

This repository is built in PyTorch 2.0.0 and tested on Ubuntu 18.04 environment (Python3.9.17, CUDA11.7). Follow these intructions

  1. Clone our repository
git clone https://github.com/akshaydudhane16/DyNet.git
cd DyNet
  1. Create conda environment The Conda environment used can be recreated using the requirements.txt file

Dataset Download and Preperation

All the 5 datasets used in the paper can be downloaded from the following locations:

Denoising: BSD400, WED, Urban100

Deraining: Train100L&Rain100L

Dehazing: RESIDE (OTS)

The training data should be placed in data/Train/{task_name} directory where task_name can be Denoise,Derain or Dehaze. After placing the training data the directory structure would be as follows:

└───Train
    ├───Dehaze
    │   ├───original
    │   └───synthetic
    ├───Denoise
    └───Derain
        ├───gt
        └───rainy

The testing data should be placed in the test directory wherein each task has a seperate directory. The test directory after setup:

├───dehaze
│   ├───input
│   └───target
├───denoise
│   ├───bsd68
│   └───urban100
└───derain
    └───Rain100L
        ├───input
        └───target