Add more CNN networks for encoding generation, allow custom model addition

✨ New features and improvements

• Add more CNN networks apart from the default MobileNetv3- the new networks added are: EfficientNet and ViT #190
• Allow user to define and plugin their own custom CNN models for encoding generation - an example of how to use this can be seen here: imagededup/examples/use_custom_model.ipynb #190
• Make encoding generation quicker and memory-efficient by removal of torch grad calculation #192

🔴 Bug fixes

• Update CNN link in readme to point to MobilenetV3 paper.
• Remove 'pretrained' parameter of mobilenetv3 to get rid of deprecation warning and use 'weights' parameter instead.
• Some typos.

v0.3.1: Add optional multiprocessing flags, gpu support, webp image support

✨ New features and improvements

• Add optional multiprocessing flags #185
• Add gpu support for CNN encoding generation #179
• Add support for webp images #153

Breaking changes: None

v0.3.0: Change CNN encodings, drop python 3.6/3.7 support

Installation fix

• Make package installable by removing tensorflow as a dependency and replacing it with pytorch #173
• Drop support for python 3.6 and python 3.7 #173

✨ New features and improvements

• Use MobileNetv3 for generating CNN encodings #173
• Introduce a 'recursive' option to generate encodings for images organized in a nested directory structure #104

Breaking changes

• Size of CNN encodings is 576 instead of 1024 #173
• Since CNN encodings are generated using a different network, the robustness might be different; user might need to change similarity threshold settings #173
• Hashes (all types) may be different from previous versions for a given image #173

v0.2.4: Fix bugs, add Python 3.8 to tested versions and small improvements

🔴 Bug fixes

• Fix broken cython brute force in Python 3.8 #117
• Close figure after plotting to avoid figure overwrite #111
• Allow encode_image method of cnn to accept 2d arrays #110
• Relax dependencies and update packages #116, #107, #102, #119

v0.2.2: Improve evaluation speed

✨ New features and improvements

• Switched to creating list comprehensions to create lists on demand instead of slower explicit for loops that rely on calling the append function in every iteration. #76
• Used sets for membership tests
• Used broadcasting instead of explicit for loops

v0.2.1: Add Manifest.in so that c files are included in the source distribution

🔴 Bug fixes

• Add Manifest.in so that c files are included in the source distribution #72

v0.2.0: Add Brute Force Cython, Expand supported image formats and some small fixes

✨ New features and improvements

• Implemented Cython implementation for brute force. This is now used as default search_method on Linux and MacOS X. For Windows, we still use bktree as default as we are not sure that popcnt is supported #56
• Expand supported image formats. Now it also supports: 'MPO', 'PPM', 'TIFF', 'GIF', 'SVG', 'PGM', 'PBM' #35

🔴 Bug fixes

• Relaxing the package dependencies #36
• Removal of print statements #39
• Fix type error when saving scores #55 & #61

👥 Contributors

Thanks to @jonatron, @orf, @DannyFeliz, @ImportTaste, @fridzema, @DannyFeliz, @iozevo, @MomIsBestFriend, @YadunandanH for the pull requests and contributions.

First release

This is the first release of imagededup.

We added:
🧮 Several hashing algorithms (PHash, DHash, WHash, AHash) and convolutional neural networks
🔎 An evaluation framework to judge the quality of deduplication
🖼 Easy plotting functionality of duplicates
⚙️ Simple API