All notable changes to this project will be documented in this file.
- Note in readme that the dataset used to create this work is available upon request by filling out this form
- Bugfix: config.py was setting the retries env variable as the like env variable; this has been corrected
- Bugfix: remove string endings from .env variable strings if they exist
- You can now store all default optional parameters in your environment variables! This means you can set your starting distance, number of likes, and image_batch size without manually specifying the options each time. You can locally set a
.envfile within your tinder working directory. Check out this .env example file. Thanks to Vikash Kothary for this contribution. - .env file will work on Windows/Linux/Mac!
- config.txt file no longer supported! Switch to a .env file, or set your environment variable!
- New single sourced version file is within setup.py.
- Fix Python 2.7 related issues with login and config.txt parsing
- retries option for tindetheus like or browse command, default to 20, and can be modified in config.txt, for example
retries = 25
- cleaned up some code for parsing config.txt, seperated interger and string types
- setup.py now lists
tensorflow < 2.0.0as a requirement
- skimage.transform.resize now uses defaults
- setuptools now shows up in the requirements
- fixed a bug that would cause tindetheus not to run if a line in config.txt was a list of 2 items when separated by a space
- switched to setuptools, remove README.rst
- Basic travis.ci checking for installation on python 2.7, 3.5, 3.6 and flake8 on files
- Reorganize tindetheus functions into submodules image_processing, machine_learning, tinder_client
- Bugfix python 2.7 installation
- Bugfix related to try except that was overriding the database folder on newer numpy distributions
- Moved installation instructions, changelog, and validate function details to separate files
- numpy.load now uses allow_pickle=True
- Remove function from tindetheus into submodules image_processing, machine_learning, tinder_client
- Fix installation on Windows
- Fix loading MTCNN weights on newer version of numpy
- Docker container instructions.
- Readme notes
- Bugfix related to Python 2.7 command line parsing
- Add option to log in using XAuthToken thanks to charlesduponpon
- Add like_folder command line option to create al/like and al/dislike folders based on the historically liked and disliked profiles (this is a quick way to assess model quality)
- Fix issue where line endings that were causing authentication failure
- Fix handling of config.txt
- New validate function to apply your tindetheus model to a new dataset. See README on how to use this function.
- Fix issues with lossy integer conversions
- Some other small bug fixes
- Update how facenet TensorFlow model is based into object
- Fixes session recursion limit
- tindetheus will now exit gracefully if you have used all of your free likes while running tindetheus like
- Fix bug related to Windows and calc_avg_emb(), which wouldn't find the unique classes
- Added version tracking and parser with --version
- New optional parameters: likes (set how many likes you have remaining default=100), and image_batch (set the number of images to load into facenet when training default=1000)
- Now all optional settings can be saved in config.txt
- Bug fix related to calling a tindetheus.export_embeddings function
- Saving the same filename in your database no longer bombs out on Windows
- Code should now follow pep8
- Added support for latest facenet models. The different facenet models don't appear to really impact the accuracy according to this post.
- You can now specify which facenet model to use in the config.txt file.
- Added example script for inspecting your database manually
- Updated facenet clone implementation
- Now requires minimum tensorflow version of 1.7.0