8 packages tagged with “RNN”
This is a lightweight package which is able to run neural networks with weights from different sources like Keras models.
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.
FULL TensorFlow 1.15 for .NET with Keras. Build, train, checkpoint, execute models. 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, including estimators. This preview will expire. !!NOTE!! This version requires Python 3.x x64 to be installed with tensorflow or tensorflow-gpu 1.15. See the official installation instructions in https://www.tensorflow.org/install/ (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 on-site/remote support for this preview email contact@losttech.software . More information in NuGet package release notes and at the project web page: https://github.com/losttech/Gradient .
Library for working with RNN, R-CNN, CNN, GAN, and classic NN for classification, generate and analyze data with in-built methods and math objects like tensors, matricies and vectors
DyNet is a neural network library developed by Carnegie Mellon University and many others. It is written in C++ (with bindings in Python and C#) and is designed to be efficient when run on either CPU or GPU, and to work well with networks that have dynamic structures that change for every training instance. For example, these kinds of networks are particularly important in natural language processing tasks, and DyNet has been used to build state-of-the-art systems for syntactic parsing (https://github.com/clab/lstm-parser), machine translation (https://github.com/neubig/lamtram), morphological inflection (https://github.com/mfaruqui/morph-trans), and many other application areas. Read the documentation (http://dynet.readthedocs.io/en/latest/) to get started, and feel free to contact the dynet-users group (https://groups.google.com/forum/#!forum/dynet-users) group with any questions (if you want to receive email make sure to select "all email" when you sign up). We greatly appreciate any bug reports and contributions, which can be made by filing an issue or making a pull request through the github page (http://github.com/clab/dynet). You can also read more technical details in our technical report (https://arxiv.org/abs/1701.03980).
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!
Welvet - LOOM Neural Network Framework for .NET. 12 layer types (Dense, LSTM, RNN, Conv2D, Conv1D, MHA, LayerNorm, RMSNorm, SwiGLU, Softmax, Embedding, Parallel), transformer/LLM inference with streaming, neural tweening, K-Means clustering, Pearson/Spearman correlation, network grafting, step-based forward pass, 7 LR schedulers, 3 optimizers (SGD/AdamW/RMSprop), ensemble features, and bit-for-bit model sharing with Python/TypeScript/Go. Multi-platform: Linux, Windows, macOS, Android, iOS.
Giving Windows C# developers easy access to the ONNX AI Model Format for easy model conversions.