This is a collection of library and scripts to download and deal with data from eumetstat. Specifically data covering the UK downloaded from the high rate SEVIRI imager from metosat 2nd gen (EO:EUM:DAT:MSG:HRSEVIRI).
A python script to download and crop images from eumetstat. It downloaded the data and extracts a 500 x 500px png file for each of the 12 layers. It might not be the best code, but it works (most of the time).
The easiest way is to clone the repo and install it. Hopefully if I have the config set up it should install all the deps
git clone https://github.com/TimCargan/eumetsat.git
cd eumetsat
pip install -e .Go to eumetsat website and get your key, put it in /scripts/eumetsat.key.
Then to run the download script, assuming the data dict is set up use:
python ./scripts/download.py --dl 2 --ep 6 --st "2015-01-01" --et "2021-01-01" --mins 0 --mins 30In addition to downloading the data there are some scripts to shape the raw pngs into a more ml friendly formats:
- tf.data
- tensorstore
- numpy memmap file