Games, NLP

A webcam-based 3x3x3 rubik’s cube solver written in Python 3

Qbr Qbr, pronounced as Cuber, is a webcam-based 3x3x3 rubik’s cube solver written inPython 3 and OpenCV. :rainbow: Accurate color detection :mag: Accurate 3x3x3 rubik’s cube detection :capital_abcd: Multilingual Solve mode Calibrate mode Isn’t the default color detection working out for you? Use the calibratemode to let Qbr be familiar with your cube’s color scheme. If your room hasproper lighting then this will give you a 99.9% guarantee that your colors willbe detected properly. Simply follow the on-screen instructions and […]

Read more
Machine Learning, NLP

Fast, differentiable sorting and ranking in PyTorch

Torchsort Pure PyTorch implementation of Fast Differentiable Sorting and Ranking (Blondel et al.). Much of the code is copied from the original Numpy implementation at google-research/fast-soft-sort, with the isotonic regression solver rewritten as a PyTorch C++ and CUDA extension. Install pip install torchsort To build the CUDA extension you will need the CUDA toolchain installed. If youwant to build in an environment without a CUDA runtime (e.g. docker), you willneed to export the environment variableTORCH_CUDA_ARCH_LIST=”Pascal;Volta;Turing” before installing. Usage torchsort exposes […]

Read more
Framework, NLP

An Agnostic Object Detection Framework

cevision End-to-End Object Detection Framework – Pluggable to any Training Library: Fastai, Pytorch-Lightning with more to come. Installation pip install icevision[all] Important: We currently only support Linux/MacOS. Why IceVision? IceVision is an Object-Detection Framework that connects to different libraries/frameworks such as Fastai, Pytorch Lightning, and Pytorch with more to come. Features a Unified Data API with out-of-the-box support for common annotation formats (COCO, VOC, etc.) The IceData repo hosts community maintained parsers and custom datasets Provides flexible model implementations with […]

Read more
NLP, Tool

A tool to convert AWS EC2 instances back and forth between On-Demand

ec2-spot-converter This tool converts existing AWS EC2 instances back and forth between On-Demand and ‘persistent’ Spot billing models while preserving instance attributes (Launch configuration, Tags..), network attributes (existing Private IP addresses, Elastic IP), storage (Volumes), Elastic Inference accelerators, Elastic GPUs. Others features: Can also perform Spot-to-Spot and OnDemand-to-OnDemand conversions: Allow replacement of existing Spot instances with new “identical” ones to update the instance type and CPU options, Help to fix some Spot instance conditions (Ex: ‘IncorrectSpotRequestState Exception’), Allow Root Disk […]

Read more
Graph, NLP

Halftone with Points And Lines

HalftonePAL This arranges points on an image for halftoning, using the circles themselves, dots at their centers, or lines connecting the circles. Features: Choose how to arrange the circles: Circle packing, quadtree, random dither, 1D and 2D error diffusion on square or hex grids. Choose what pattern to make from the circles: Show dots, circles, tilted lines, scan back-and-forth, connect the points with a Hilbert curve, make a greedy path between points, a greedy loop, a minimum spanning tree, a […]

Read more
Machine Learning, NLP

Focal Frequency Loss for Generative Models

focal-frequency-loss This repository will provide the official code for the following paper: Focal Frequency Loss for Generative ModelsLiming Jiang, Bo Dai, Wayne Wu and Chen Change LoyarXiv preprint, 2020. Abstract: Despite the remarkable success of generative models in creating photorealistic images using deep neural networks, gaps could still exist between the real and generated images, especially in the frequency domain. In this study, we find that narrowing the frequency domain gap can ameliorate the image synthesis quality further. To this […]

Read more
Machine Learning, NLP

Writing high-level code for parallel high-performance stencil computations

ParallelStencil.jl ParallelStencil empowers domain scientists to write architecture-agnostic high-level code for parallel high-performance stencil computations on GPUs and CPUs. Performance similar to CUDA C can be achieved, which is typically a large improvement over the performance reached when using only CUDA.jl Array programming. For example, a 2-D shallow ice solver presented at JuliaCon 2020 [1] achieved a nearly 20 times better performance than a corresponding CUDA.jl Array programming implementation; in absolute terms, it reached 70% of the theoretical upper performance […]

Read more
NLP

Machine Learning and Deep Learning Courses

If you follow me on Twitter, you will know that I have been curating and sharing some of the most recent machine learning, natural language processing (NLP), and deep learning courses that have been publicly released. I have completed the majority of these courses and some I am currently reviewing. At the request of some people, I have decided to write this blog post so that it is easier to maintain and bookmark the complete collection of courses. Here they […]

Read more
Database, NLP

Python Client for TerminusDB

TerminusDB Client Python Python version of the TerminusDB client – for TerminusDB API and WOQLpy Requirements Installation TerminusDB Client can be downloaded form PyPI using pip:python -m pip install terminusdb-client This only includes the core Python Client (WOQLClient) and WOQLQuery. If you want to use woqlDataframe: python -m pip install terminusdb-client[dataframe] if you are installing form zsh you have to quote the argument like this: python -m pip install ‘terminusdb-client[dataframe]’ python -m pip install git+https://github.com/terminusdb/terminusdb-client-python.git Usage >>> from terminusdb_client import […]

Read more
Machine Learning, NLP

Detectron2 for Document Layout Analysis

Detectron2 This repo contains the training configurations, code and trained models trained on PubLayNet dataset using Detectron2 implementation.PubLayNet is a very large dataset for document layout analysis (document segmentation). It can be used to trained semantic segmentation/Object detection models. NOTE Models are trained on a portion of the dataset (train-0.zip, train-1.zip, train-2.zip, train-3.zip) Trained on total 191,832 images Models are evaluated on dev.zip (~11,000 images) Backbone pretrained on COCO dataset is used but trained from scratch on PubLayNet dataset Trained […]

Read more
1 … 701 702 703 704 705 … 947