Supporting library for Machine Learning capabilities in the Transformation.Formula DSL.
A complete C# re-write of Berkeley's open source Convolutional Architecture for Fast Feature Encoding (CAFFE) for Windows C# Developers with full On-line Help, now with Temporal Fusion Transformers, GPT, Seq2Seq/Attention, Single-Shot MultiBox, TripletNet, SiameseNet, NoisyNet, Deep Q-Network and Policy Gradient Reinforcement Learning, cuDNN LSTM Recurrent Learning, and Neural Style Transfer support!
This package contains the default German models for Catalyst. Catalyst is a Natural Language Processing library built from scratch for speed. Inspired by spaCy"s design, it brings pre-trained models, out-of-the box support for training word and document embeddings, and flexible entity recognition models.
Microsoft.ML.Experimental contains experimental work such extension methods to access internal methods.
The wrapper provides a simple and easy-to-use interface for loading the MiniLM embeddings model and generating embeddings for input text.
Integration for Catalyst to run spaCy models from C#.
The Microsoft.ML.Tokenizers.Data.R50kBase includes the Tiktoken tokenizer data file r50k_base.tiktoken, which is utilized by models such as text-davinci-001
Extreme Optimization Numerical Libraries for .NET Data Access Library for JSON.
IBM.WatsonDeveloperCloud.Conversation.v1 wraps the Watson Developer Cloud Conversation service (http://www.ibm.com/watson/developercloud/conversation.html)
This is a .NET wrapper for the FastText library by Facebook AI Research (FAIR). This package facilitates the integration of language identification capabilities into .NET applications, leveraging the powerful and efficient FastText library to recognize and classify texts in multiple languages. Ideal for applications that require fast and accurate language detection.
WeightedSelector.NET lets you assign weights to a set of choices, then make intelligent decisions based on each choice's proportion of the total weight. This simple concept is useful for machine learning scenarios where choices need to be made based on dynamic factors. Great examples include suggestion engines and game AI. In one of the examples, we implement a game AI that decides between attacking, fleeing, and casting a heal spell based on a dynamic "fear" factor... In 6 lines of code. Free to use under MIT License.
This package provides the native resources required for the FastText library by Facebook AI Research (FAIR). It includes the necessary native components to enable language recognition and classification capabilities in applications using the FastTextWrapper for .NET. This package is essential for the proper functioning of FastText-based language identification features in your .NET projects.
TorchSharp makes PyTorch available for .NET users. libtorch-cuda-11.1-linux-x64 contains components of the PyTorch LibTorch library version 1.9.0 redistributed as a NuGet package with added support for TorchSharp.
IBM.WatsonDeveloperCloud.SpeechToText.v1 wraps the Watson Developer Cloud Speech To Text service (http://www.ibm.com/watson/developercloud/speech-to-text.html)
Contains the IDataView system which is a set of interfaces and components that provide efficient, compositional processing of schematized data for machine learning and advanced analytics applications.
Provides Models for the NuvAI Platform.
FULL TensorFlow 1.15 for .NET with Keras. Build, train, checkpoint, execute models. Samples: https://github.com/losttech/Gradient-Samples, https://github.com/losttech/YOLOv4, https://github.com/losttech/Siren Deep learning with .NET blog: https://ml.blogs.losttech.software/ Comparison with TensorFlowSharp: https://github.com/losttech/Gradient/#why-not-tensorflowsharp Comparison with TensorFlow.NET: https://github.com/losttech/Gradient/#why-not-tensorflow-net Allows building arbitrary machine learning models, training them, and loading and executing pre-trained models using the most popular machine learning framework out there: TensorFlow. All from your favorite comfy .NET language. Supports both CPU and GPU training (the later requires CUDA or a special build of TensorFlow). Provides access to full tf.keras and tf.contrib APIs, estimators and many more. Free for non-commercial use. For licensing options see https://losttech.software/buy_gradient.html !!NOTE!! This version requires Python 3.x x64 to be installed with tensorflow 1.15.x. See the official installation instructions in https://www.tensorflow.org/install/pip#older-versions-of-tensorflow (ensure you are installing version 1.15 to avoid hard-to-debug issues). Please, report any issues to https://github.com/losttech/Gradient/issues For community support use https://stackoverflow.com/ with tags (must be all 3 together) tensorflow, gradient, and .net. For support email contact@losttech.software . More information in NuGet package release notes and on the project web page: https://github.com/losttech/Gradient . TensorFlow, the TensorFlow logo and any related marks are trademarks of Google Inc.
IBM.WatsonDeveloperCloud.VisualRecognition.v3 wraps the Watson Developer Cloud Visual Recognition service (http://www.ibm.com/watson/developercloud/visual-recognition.html)
Dew.Math is the Windows-optimized high-performance numerical computation library for .NET. It provides a vectorized matrix and vector math environment with native runtime acceleration, multithreaded execution, and extensive algorithm libraries for scientific, engineering, financial, AI/ML and signal processing workloads. Core Numerical Capabilities: - Dense linear algebra (BLAS, LAPACK): SVD, QR, LQ, LU, eigenvalue problems, least-squares, rank reveals - Sparse matrix support: direct solvers (Pardiso, UMFPACK), iterative solvers (CG, BiCG, GMRES), preconditioning strategies, structured sparse formats - Complex number computation with fully vectorized math operations - Polynomial arithmetic, interpolation, splines, rational approximations, Chebyshev basis transforms - Numerical differentiation, root solving, non-linear systems, ODE support for stiff and non-stiff cases - Probability distributions (over 30 families), random number generators, Monte Carlo methods - Special mathematical functions (Airy, Bessel, Gamma-related, elliptic integrals, Legendre, etc.) Optimization and Modeling: - Non-linear curve fitting with Levenberg-Marquardt and trust-region refinements - Direct and constrained optimization (Simplex/Nelder–Mead, BFGS, Conjugate Gradient, LP, dual-phase simplex, Gomory cutting plane) - Vectorized expression parser for dynamic formula construction and symbolic-style evaluation Performance Architecture: - Native accelerated BLAS/LAPACK kernels with automatic CPU dispatch (AVX, AVX2, AVX-512) - Scalable multithreading with a lock-free memory allocator for low-GC overhead - Optional OpenCL GPU offloading for supported device targets Platform Model: - Contains Windows native acceleration binaries - For Linux native acceleration use: Dew.Math.Linux - For a pure managed, portable edition use: Dew.Math.Core Use Dew.Math when you require **maximum numerical performance on Windows** for HPC, simulation, economic modeling, data analytics, or scientific visualization workflows.
A comprehensive .NET library for machine learning, deep learning, NLP, computer vision, and AI model serving. Licensed under BSL 1.1 — free for non-commercial use, community license available at aidotnet.dev. Model save/load requires a free or paid license after a 10-operation trial. Optional anonymous telemetry (opt-in via AIDOTNET_TELEMETRY=true) collects usage metrics — no PII or model data is collected.
DiffSharp is an automatic differentiation (AD) library. AD allows exact and efficient calculation of derivatives, by systematically invoking the chain rule of calculus at the elementary operator level during program execution. AD is different from numerical differentiation, which is prone to truncation and round-off errors, and symbolic differentiation, which is affected by expression swell and cannot fully handle algorithmic control flow. Using the DiffSharp library, derivative calculations (gradients, Hessians, Jacobians, directional derivatives, and matrix-free Hessian- and Jacobian-vector products) can be incorporated with minimal change into existing algorithms. Diffsharp supports nested forward and reverse AD up to any level, meaning that you can compute exact higher-order derivatives or differentiate functions that are internally making use of differentiation. Please see the API Overview page for a list of available operations. The library is under active development by Atılım Güneş Baydin and Barak A. Pearlmutter mainly for research applications in machine learning, as part of their work at the Brain and Computation Lab, Hamilton Institute, National University of Ireland Maynooth. DiffSharp is implemented in the F# language and can be used from C# and the other languages running on Mono or the .Net Framework, targeting the 64 bit platform. It is tested on Linux and Windows. We are working on interfaces/ports to other languages.
Welcome to Sailfish! There are three tools included in this package: - Sailfish is a simple in-process performance test framework for writing clean, manageable performance tests. - SailDiff is a statistical testing tool for comparing before change and after change datasets. - ScaleFish is a machine learning tool for modeling general logical complexity and making predictions. Use these in combination to write a basic performance regression detection system for your application's delivery pipeline.
Microsoft Machine Learning Scoring library for deep learning model inference. Current version of the library supports inferencing on ONNX v1.3 and TensorFlow v1.10.0 models. The library supports CPU execution with MKL/MKLDNN acceleration. Also supports CUDA GPU devices. For CPU execution of ONNX models, no extra libraries are required. However for scoring TensorFlow models, the CUDA libraries are needed for both CPU and GPU execution. Download and install CUDA 9.2 toolkit, CUDNN and device drivers separately. This package provides a .Net standard 1.3 compatible module for maximum portability. Currently supported platforms include x64 CPU on Windows OS only.
TorchSharp makes PyTorch available for .NET users. libtorch-cuda-11.7-win-x64 contains components of the PyTorch LibTorch library version 2.0.1 redistributed as a NuGet package with added support for TorchSharp.
IBM.WatsonDeveloperCloud.ToneAnalyzer.v3 wraps the Watson Developer Cloud Tone Analyzer service (http://www.ibm.com/watson/developercloud/tone-analyzer.html)
The gRPC C# client for the Clarifai API.
TorchSharp makes PyTorch available for .NET users. libtorch-cuda-11.3-linux-x64 contains components of the PyTorch LibTorch library version 1.11.0 redistributed as a NuGet package with added support for TorchSharp.
IBM.WatsonDeveloperCloud.TextToSpeech.v1 wraps the Watson Developer Cloud Text To Speech service (http://www.ibm.com/watson/developercloud/text-to-speech.html)
Provides Interfaces, Models and Managers for NuvAI Platform.
Contains classes to download and parse machine learning datasets such as MNIST, News20, Iris. This package is part of the Accord.NET Framework.
Aspose. OCR Cloud is a simple ocr technology, which you can use in your application to convert image to text. Add OCR to your app with simply few line of code. Any language, any platform, no hardware resources. Automatic skew correction and advanced automative image pre-processing: Aspose.OCR Cloud can successfully read poor quality scans and automatically make image pre-processing to correct low resolution, low contrast, noise and skew, image upscale resolution. 45 languages supported including Chinese. Automated and manual document layout detection: Recognize documents with complex layouts in fully automatic mode or with manual corrections. In automated mode, you don’t need to worry about text layout editing. Our DSR algorithm provides such functionality. It allows performing the OCR operation quickly on various document scans. Receipt recognition. Text region detection method Table image recognition. PDF recognition. Multipage supported. Text correction using spell checkong algorithms. Various Output formats: Text, Searchable PDF, hOCR. Supported image formats: We support all widely used image formats: bmp jpeg jpg jpe jp2 png webp tiff tif exr hdr pic svg Recognize Text of Different Languages: Supports 45 languages including: English, German, French, Italian, Spanish, Portuguese, Polish, Slovene, Slovak, Netherlands, Lithuanian, Latvian, Danish, Norwegian, Finnish, Serbian, Croatian, Czech, Swedish, Estonian, Romanian, Chinese, Arabic, Hindi, Russian, Ukrainian, Bengali, Tibetan, Thai, Urdu, Turkish, Korean, Indonesian, Hebrew, Javanese, Greek, Japanese, Persian, Albanian, Latin, Vietnamese, Uzbek, Georgian, Bulgarian, Azerbaijani Recognize english handwritten text. High speed with no hardware resources: Don’t worry about hardware, all the computational operations are performed at our powerful servers, that hosts our core, built on cutting edge fast neural networks.
ConceptNet extension for Catalyst. You need to install your language packages as well, such as Catalyst.ConceptNet.English
Infer.NET is a framework for running Bayesian inference in graphical models. It can also be used for probabilistic programming. This package contains complete machine learning applications including a classifier and a recommender system.
IBM.WatsonDeveloperCloud.PersonalityInsights.v3 wraps the Watson Developer Cloud Personality Insights service (http://www.ibm.com/watson/developercloud/personality-insights.html)
IBM.WatsonDeveloperCloud.NaturalLanguageUnderstanding.v1 wraps the Watson Developer Cloud Natural Language Understanding service (http://www.ibm.com/watson/developercloud/natural-language-understanding.html)
.NET for Android and MAUI bindings for the Android Java library 'org.tensorflow:tensorflow-lite-api'. Library description: A library helps deploy machine learning models on mobile devices
Bright Wire is an open source machine learning library. Includes neural networks (feed forward, convolutional and recurrent), naive bayes, linear regression, decision trees, logistic regression, k-means clustering and dimensionality reduction.
Microsoft.ML.CpuMath contains optimized math routines for ML.NET.
IBM.Watson.Assistant.v1 wraps the Watson Assistant service (http://www.ibm.com/watson/developercloud/assistant.html)
The Collection of the Machine Learning related functionality for AspNetCore or GenericHost.
ML.NET is a cross-platform open-source machine learning framework which makes machine learning accessible to .NET developers.
ML.NET additional learners making use of Intel® oneAPI Data Analytics Library (oneDAL).