Downloading sky image datasets for pytorch applications
Project description
skyimages
Download sky images for solar power forecasting as pytorch-ready datasets.
Installation using pypi
From release v0.0.1, skyimages
will be installable via pip:
pip install skyimages
Installation using github
Download skyimages
from
github.
Change the CWD to the download folder skyimages/
and install
the package using pip.
git clone git@github.com:FlorianK13/skyimages.git
cd skyimages
pip install .
Getting started
To download the dataset SKIPP'D (source) and use it as a pytorch torchvision dataset, run the following code in python
from skyimages.dataset import SKIPPDDataSet
import torch
traindata = SKIPPDDataSet(download=True)
testdata = SKIPPDDataSet(download=True, train=False)
trainloader = torch.utils.data.DataLoader(
traindata, batch_size=200, shuffle=True, num_workers=0
)
testloader = torch.utils.data.DataLoader(
testdata, batch_size=20, shuffle=False, num_workers=0
)
Running code on Windows
If you use windows, wrap your code in the if __name__ == '__main__'
clause to support multiprocessing.
Here you can find more info on this issue.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
skyimages-0.0.2.tar.gz
(7.5 kB
view hashes)