Python toolbox based on Python 3, PyTorch and OpenCV-Python.
📓 [Wiki]
📧 Feel free to contact: ryanxingql@gmail.com
.
- v2: add new functions; improve efficiency; not compatible with v1.
- v1: the first formal version.
- algorithm: a basis class of deep-learning algorithms.
- conversion: common-used tools for format/color space conversion, e.g., YCbCr <-> RGB conversion.
- dataset: basic datasets and common-used functions, e.g., LMDB dataset.
- deep learning: common-used tools for deep learning, e.g., multi-processing.
- metrics: common-used metrics, e.g., PSNR.
- network: a basis class of deep-learning networks.
- system: common-used tools for system manipulation, e.g., return formatted time data.
- In default, an image is a NUMPY array with the size being
(H W C)
and the data type beingnp.uint8
. - In default, an image input is first copied then processed; a tensor input is
[.cpu()].detach().clone()
then processed.
We adopt Apache License v2.0. For other licenses, please refer to BasicSR.
If you find this repository helpful, you may cite:
@misc{2021xing,
author = {Qunliang Xing},
title = {Python Utils},
howpublished = "\url{https://github.com/ryanxingql/pythonutils}",
year = {2021},
note = "[Online; accessed 11-April-2021]"
}