A general-purpose Dataloader for Tensorflow 2.x. It supports many medical image formats.
Project description
Medical Images Dataloader
A general-purpose Dataloader for Tensorflow 2.x. It supports many medical image formats.
Free software: MIT license
Documentation: https://med-dataloader.readthedocs.io.
Features
TODO
Credits
This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.
History
0.1.8 (2021-05-09)
Main Changes in the package structure. Now there are two main functions: generate_dataset and get_dataset, both leveraging on DataLoader class.
The generation of the dataset can be handled also by CLI, to simplify usage.
Processed data can live by themself. No more need to transfer also original file (e.g. to Drive to make use of them on Colab)
0.1.6 (2021-05-06)
Improved flexibility for image data types. Now cache dimension reflects the actual dataset dimension.
0.1.5 (2021-04-30)
Added support for 3D files: now Dataloader automatically detects whether a file is 2D or 3D and returns the properly sized dataset. Please remember that med_dataloader returns tf.data.Dataset object for 2D tasks, 3D is not yet supported.
Added new notebook in examples folder.
0.1.4 (2021-04-29)
- Improved code flexibility:
It is possibile to choose which type of data augmentation is performed
Boundaries for data normalization can be set by the user
Images can be resized automatically by the user
Added basic_usage example also as a notebook
0.1.1 (2021-04-20)
Added code for package
Basic example of usage inside folder “examples”
Partial documentation
0.1.0 (2021-04-16)
First release on PyPI.
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
Built Distribution
Hashes for med_dataloader-0.1.8-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 120266da59ecd64b5ec6651a52929c7f50da47a812d057331cae9cb0127203f6 |
|
MD5 | 0661bcb5d56bb889c4dda12e6408b51b |
|
BLAKE2b-256 | 7951975bfe40e7426bffecd37a493b8b9014720fd65ae5eea38b666908c01121 |