Climate Reconstruction via Image Inpainting using Partial Convolutions

Applied implementation

Official implementation is released by the authors.

Climate reconstruction implementation

This is an unofficial pytorch implementation of a paper, Image Inpainting for Irregular Holes Using Partial Convolutions [Liu+, arXiv2018].

Requirements

• Python 3.6+
• Pytorch 0.4.1+

pip install -r requirements.txt

Usage

Preprocess

• download climate data and preprocess it to hdf5 (square image, 72x72 is default) (see mask dir). Use anomalies (image normalization is turned off). The dataset should contain data_large (training data), val_large, and test_large as the subdirectories. Don't forget to specify the root of the dataset by --root ROOT when using train.py or test.py

Use the start.sh script to operate the process chain: Train, Fine-Tune, Test