Deepsaki

Latest version: v1.0.1

Safety actively analyzes 681812 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

1.0.1

What's Changed
* Update documentation by sascha-kirch in https://github.com/sascha-kirch/DeepSaki/pull/26
* Update logo font. add social card. added meta tags. by sascha-kirch in https://github.com/sascha-kirch/DeepSaki/pull/29
* Add dev status identifier by sascha-kirch in https://github.com/sascha-kirch/DeepSaki/pull/33
* Increase version to 1.0.1. Changed refferences for social cards. by sascha-kirch in https://github.com/sascha-kirch/DeepSaki/pull/34
* Update changelog. by sascha-kirch in https://github.com/sascha-kirch/DeepSaki/pull/37


**Full Changelog**: https://github.com/sascha-kirch/DeepSaki/compare/v1.0.0...v1.0.1

1.0.0

<p align="center">
<img src="https://github.com/sascha-kirch/DeepSaki/blob/main/docs/assets/images/ds_logo.png" alt="deepsaki-logo" width = 80% >
</p>


**Changes:**

- **Framework:**
- Upgrade from python 3.8 to python 3.10 since it supports a wide range of tensorflow version(2.6-2.14)
- update of project structure to a modern SW project
- automated CI/CD
- docstrings and type annotations
- initial automated documentation using mkdocs and mike supporting multiple version. Will be further improved in a future.
- automated code coverage reports
- automated publishing of packages on release
- add changelog and contribution documentation
- providing a dockerfile for the dev environment

- **DeepSaki.initializers:**
- renamed DeepSaki.initializer -> DeepSaki.initializers
- Refactored function make_initializer_complex() into a class called ComplexInitializer

- **DeepSaki.layers:**
- renamed ResidualIdentityBlock -> ResidualBlock
- New Layers:
- LearnedPooling
- FFT3D
- iFFT3D

- **DeepSaki.losses:**
- New abstract base class for image based losses.

- **DeepSaki.optimizers:**
- renamed module from optimizer -> optimizers
- new method switch_optimizer() to change the current optimizer of SwatsAdam or SwatsNadam optimizer.

- **DeepSaki.augmentations:**
- renamed module from regularizations to augmentations since it only contained augmentations.

- **DeepSaki.utils:**
- renamed DetectHw() -> detect_accelerator()
- renamed EnableXlaAcceleration() -> enable_xla_acceleration()
- renamed EnableMixedPrecision() -> enable_mixed_precision()

- **New modules:**
- DeepSaki.types: provides type definitions and enums used in the package.

- **Misc:**
- general code cleaning and refactoring.
- changed print expressions to logging
- using pyproject.toml for project config

- **Outlook:**
- improved changelogs with issue tickets
- new modules are planned, e.g. a diffusion module and a framework module featuring CycleGAN and DDPM.

0.1.3

Fourier Convolution does now consider circular convolution

**Full Changelog**: https://github.com/sascha-kirch/DeepSaki/compare/v0.1.2...v0.1.3

0.1.2

Release Note
- Added layers working with fourier transformations
- code cleaning

0.1.0

Links

Releases

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.