Top 21 Python unsupervised-learning Projects
(JMLR'19) A Python Toolbox for Scalable Outlier Detection (Anomaly Detection)Project mention: PyOD: ~50 anomaly detection algorithms in one framework. | reddit.com/r/algotrading | 2021-01-25
The standard package for machine learning with noisy labels and finding mislabeled data. Works with most datasets and models. (by cleanlab)Project mention: [Discussion] What is your go to technique for labelling data? | reddit.com/r/MachineLearning | 2021-09-15
You can save a lot of money using cleanlab: https://github.com/cleanlab/cleanlab
Run Linux Software Faster and Safer than Linux with Unikernels.
The standard package for machine learning with noisy labels and finding mislabeled data. Works with most datasets and models.Project mention: [P] Confident Learning making ML QA 34x cheaper | reddit.com/r/MachineLearning | 2021-08-14
Code for https://arxiv.org/abs/1911.00068 found: https://github.com/cgnorthcutt/cleanlab
Karate Club: An API Oriented Open-source Python Framework for Unsupervised Learning on Graphs (CIKM 2020)
Composable GAN framework with api and user interfaceProject mention: So I trained an AI to generate Pokemon sprites and this is the result | reddit.com/r/teenagers | 2021-01-21
There is something called HyperGAN which builds generative adversarial networks (GANs) and those networks take some images as input and give those as output. Here is the GitHub page for that.
:red_circle: MiniSom is a minimalistic implementation of the Self Organizing Maps
Algorithms for outlier, adversarial and drift detectionProject mention: [D] Is this a reasonable assumption in machine learning? | reddit.com/r/MachineLearning | 2021-07-05
All of the above functionality and more can be easily used under a simple API in https://github.com/SeldonIO/alibi-detect.
Scout APM: A developer's best friend. Try free for 14-days. Scout APM uses tracing logic that ties bottlenecks to source code so you know the exact line of code causing performance issues and can get back to building a great product faster.
SCAN: Learning to Classify Images without Labels, incl. SimCLR. [ECCV 2020]Project mention: Any reference or idea about how to train unsupervised CNN model ? | reddit.com/r/deeplearning | 2021-04-13
an open-source implementation of sequence-to-sequence based speech processing engine (by athena-team)
Hierarchical unsupervised and semi-supervised topic models for sparse count data with CorExProject mention: Trying to read text documents and allow for up to m labels per documents, like suggested tags, but the number of labels can be different for each document. Any advice? | reddit.com/r/learnmachinelearning | 2021-04-11
Unsupervised is also possible for topic modelling: CorEX
Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals. [ICCV 2021]Project mention: Unsupervised semantic segmentation | reddit.com/r/MLQuestions | 2021-09-09
Check out these unsupervised masks created in exactly such way in this paper. They are nearly perfect
Code for the paper "Training GANs with Stronger Augmentations via Contrastive Discriminator" (ICLR 2021)Project mention: [D] What is the smallest dataset you styleGAN2 trained? | reddit.com/r/MachineLearning | 2021-06-04
Well, I've been trying to train a 1024 GAN from scratch on stylegan2-ada-pytorch with a small dataset 300 samples of not so diversity in images of painting faces. Fact is that on first try FID went as low as 71 and started deteriorating. Now I x-flip augmented the dataset (700 images) and at 900kimg FID went 64 but I doubt it will get lower. I lowered the learning rate to 0.0001 as they say it might help... Recently found this way of dataset augmentation... probably will use this https://github.com/jh-jeong/ContraD
🐍 Python Implementation and Extension of RDF2VecProject mention: [P] pyRDF2Vec 0.2.0 is out! | reddit.com/r/MachineLearning | 2021-03-22
This release is packed with many new features and optimizations under the hood. An entire overview of what's new can be found in our CHANGELOG (https://github.com/IBCNServices/pyRDF2Vec/releases/tag/0.2.0). An overview of some major updates:
Official implementation of the paper "DETReg: Unsupervised Pretraining with Region Priors for Object Detection".Project mention: Researchers From Tel Aviv University, UC Berkeley and NVIDIA Introduce ‘DETReg’, A Novel Unsupervised AI For Object Detection | reddit.com/r/computervision | 2021-08-01
Tensorflow implementation of Unsupervised Depth Completion from Visual Inertial Odometry (in RA-L January 2020 & ICRA 2020)Project mention: Unsupervised Depth Completion from Visual Inertial Odometry | news.ycombinator.com | 2021-08-30
Hey there, interested in camera and range sensor fusion for point cloud (depth) completion?
Here is an extended version of our [talk](https://www.youtube.com/watch?v=oBCKO4TH5y0) at ICRA 2020 where we do a step by step walkthrough of our paper Unsupervised Depth Completion from Visual Inertial Odometry (joint work with Fei Xiaohan, Stephanie Tsuei, and Stefano Soatto).
In this talk, we present an unsupervised method (no need for human supervision/annotations) for learning to recover dense point clouds from images, captured by cameras, and sparse point clouds, produced by lidar or tracked by visual inertial odometry (VIO) systems. To illustrate what I mean, here is an [example](https://github.com/alexklwong/unsupervised-depth-completion-visual-inertial-odometry/blob/master/figures/void_teaser.gif?raw=true) of the point clouds produced by our method.
Our method is light-weight (so you can run it on your computer!) and is built on top of [XIVO] (https://github.com/ucla-vision/xivo) our VIO system.
For those interested here are links to the [paper](https://arxiv.org/pdf/1905.08616.pdf), [code](https://github.com/alexklwong/unsupervised-depth-completion-visual-inertial-odometry) and the [dataset](https://github.com/alexklwong/void-dataset) we collected.
Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistency (AAAI 2021)Project mention: Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistency | reddit.com/r/depthMaps | 2021-02-09
The official PyTorch code is available here: https://github.com/SeokjuLee/Insta-DM
Student–Teacher Anomaly Detection with Discriminative Latent EmbeddingsProject mention: [R] Introduction to Fast Dense Feature Extraction -- A fast way to extract visual features for many patches from an image | reddit.com/r/MachineLearning | 2021-07-31
Code for https://arxiv.org/abs/1911.02357 found: https://github.com/denguir/student-teacher-anomaly-detection
Pipeline of a keylogging attack using just an audio signal and unsupervised learning.Project mention: Sound-based keylogging: Clustering keystroke audio recordings with t-SNE | news.ycombinator.com | 2021-01-10
SuSi: Python package for unsupervised, supervised and semi-supervised self-organizing maps (SOM)
Revisiting Contrastive Methods for Unsupervised Learning of Visual Representations. [NeurIPS 2021]Project mention: [R] Contrastive Visual Representation Learning Is More Robust Than You Might Think (Paper + Analysis) | reddit.com/r/MachineLearning | 2021-06-17
Implementation of ProteinBERT in PytorchProject mention: [R] ProteinBERT: A universal deep-learning model of protein sequence and function | reddit.com/r/MachineLearning | 2021-06-01
What are some of the best open-source unsupervised-learning projects in Python? This list will help you:
Are you hiring? Post a new remote job listing for free.