S2-BNN
lightly
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
S2-BNN
-
[R] Knowledge Distillation, Model Ensemble and Its Application on Visual Recognition - Link to a free online lecture by the author in comments
CVPR 2021 paper ‘S2-BNN: Bridging the Gap Between Self-Supervised Real and 1-bit Neural Networks via Guided Distribution Calibration’, Github: https://github.com/szq0214/S2-BNN
lightly
- Show HN: Lightly – A Python library for self-supervised learning on images
- GitHub - lightly-ai/lightly: A python library for self-supervised learning on images.
- A Python library for self-supervised learning on images
-
[P] Release of lightly 1.2.39 - A python library for self-supervised learning
Another year of has passed, and we’ve seen exciting progress in research around self-supervised learning in computer vision. We’re very excited that some of the recent models such as Masked Autoencoders (MAE) or Masked Siamese Networks (MSN) have been added to our OSS framework.
-
Self-Supervised Models are More Robust and Fair
If you’re interested in self-supervised learning and want to try it out yourself you can check out our open-source repository for self-supervised learning.
-
[D] Can a Siamese Neural Network work for invoice classification?
I assume that you have an image of the invoice. Then using a framework like https://github.com/lightly-ai/lightly with many implemented algorithms is the way to go. And after that step, with model-producing embeddings, you need to compare the embedding of a query with your known database and check if the distance is below some threshold. Of course, pipeline with checking the closest neighbor can be more complicated but I would start with sth really simple.
-
[P] TensorFlow Similarity now self-supervised training
https://github.com/lightly-ai/lightly implements a lot of self supervised models, and had been available for a while.
-
Launch HN: Lightly (YC S21): Label only the data which improves your ML model
modAL indeed has a similar goal of choosing the best subset of data to be labeled. However it has some notable differences:
modAL is built on scikit-learn which is also evident from the suggested workflow. Lightly on the other hand was specifically built for deep learning applications supporting active learning for classification but also object detection and semantic segmentation.
modAL provides uncertainty-based active learning. However, it has been shown that uncertainty-based AL fails at batch-wise AL for vision datasets and CNNs, see https://arxiv.org/abs/1708.00489. Furthermore it only works with an initially trained model and thus labeled dataset. Lightly offers self-supervised learning to learn high dimensional embeddings through its open-source package https://github.com/lightly-ai/lightly. They can be used through our API to choose a diverse subset. Optionally, this sampling can be combined with uncertainty-based AL.
- Lightly – A Python library for self-supervised learning on images
-
Active Learning using Detectron2
You can easily train, embed, and upload a dataset using the lightly Python package. First, we need to install the package. We recommend using pip for this. Make sure you're in a Python3.6+ environment. If you're on Windows you should create a conda environment.
What are some alternatives?
temporal-shift-module - [ICCV 2019] TSM: Temporal Shift Module for Efficient Video Understanding
pytorch-metric-learning - The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.
MEAL-V2 - MEAL V2: Boosting Vanilla ResNet-50 to 80%+ Top-1 Accuracy on ImageNet without Tricks. In NeurIPS 2020 workshop.
simsiam-cifar10 - Code to train the SimSiam model on cifar10 using PyTorch
byol - Implementation of the BYOL paper.
comma10k - 10k crowdsourced images for training segnets
dino - PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO
byol-pytorch - Usable Implementation of "Bootstrap Your Own Latent" self-supervised learning, from Deepmind, in Pytorch
DataProfiler - What's in your data? Extract schema, statistics and entities from datasets
Transformer-SSL - This is an official implementation for "Self-Supervised Learning with Swin Transformers".
Ne2Ne-Image-Denoising - Deep Unsupervised Image Denoising, based on Neighbour2Neighbour training
modAL - A modular active learning framework for Python