New stable release of DeepTrack2 1.5.6!
https://t.co/bCtAKIA3WB
PyPi 1.5.6
https://t.co/8BSHpN3Qti
DeepTrack2 1.5.6 is a patch that fixed an issue with the endothelial_vs dataset and an issue with repeated oneof-features.
New stable release of DeepTrack2 1.5.2!
https://t.co/m2ZvWNK1IK
PyPi 1.5.2
https://t.co/WZdHXeLwT4
DeepTrack2 1.5.2 is a patch with improvements to attention layers, new datasets to train GNNs and virtual staining NN, and improved Mie optics.
LodeSTART. A new #deeplearning method based on DeepTrack is now published in @NatureComms
It detects microscopic objects with sub-pixel accuracy from a single unlabeled image by exploiting roto-translational symmetries
Our new method to to train neural networks with a single image and without ground truth, @DeepTrackAI LodeSTAR, is published in @NatureComms
Find out more https://t.co/Fb1pu5e1EW
An effort led by @BMidtvedt
New stable release of DeepTrack2 1.5.1!
https://t.co/jFytHVnKJ1
PyPi 1.5.1
https://t.co/hYQFPqPPZb
DeepTrack2 1.5.1 is a patch with some fixes to the Mie scatterers.
DeepTrack2 1.5.0 includes:
- a new time-distributed convolutional neural network model (TimeDistributedFullyConvolutional)
- multi-input support for continuous generators
- several improvements to the Mie physics simulation code
- many other not-so-exciting bug fixes
2/2
There are two breaking changes:
1. The deprecated operator + to chain features has been removed. It is now only possible using the >> operator.
2. The deprecated operator ** to duplicate a feature has been removed. It is now only possible using the ^ operator.
5/5
DeepTrack 2.1.0.0 is now released!
https://t.co/udwg2lY58m
PyPI version 1.0.0
https://t.co/AhWkVcHyHr
DeepTrack 2.1 is a major release focusing on enhancing performance, stability, and the convenience of using DeepTrack.
1/5
If you are already using DeepTrack 2.0 (pypi version 0.x.x), updating to DeepTrack 2.1 (pypi version 1.0.0) should be painless, as no change to your code is needed, if you have followed deprecation warnings.
4/5