Microsoft Cognitive Toolkit
Microsoft Cognitive Toolkit
An open-source toolkit for distributed deep learning with support for popular model types and ONNX format.
Pricing
Free
New Features
APIOpen Source
Tool Info
Rating: N/A (0 reviews)
Date Added: October 26, 2023
Categories
GitHub
Description
Microsoft Cognitive Toolkit (CNTK) is an open-source toolkit for distributed deep learning. It supports popular model types and can be included as a library in Python, C#, or C++ programs. CNTK also supports the Open Neural Network Exchange ONNX format. The latest release of CNTK supports ONNX v1.0.
Key Features
- Implements popular network architectures
- Distributed training across GPUs/servers
- Included as a library in Python, C#, C++
- Standalone use through BrainScript
- Supports Linux and Windows OS
- ONNX model interoperability
Use Cases
- Deep Learning Development CNTK can be used as a library in Python, C#, or C++ programs for developing deep learning models. It allows users to easily realize and combine popular model types such as feed-forward DNNs, convolutional neural networks (CNNs), and recurrent neural networks (RNNs/LSTMs).
- Standalone Machine Learning Tool CNTK can also be used as a standalone machine-learning tool through its model description language, BrainScript. This makes it a useful tool for those who want to develop machine learning models without having to write code in Python, C#, or C++.
- Model Evaluation The CNTK model evaluation functionality can be used from Java programs. This makes it a useful tool for those who want to evaluate their deep learning models using Java.
- ONNX Support CNTK is one of the first deep-learning toolkits to support the Open Neural Network Exchange ONNX format. This makes it a useful tool for those who want to develop models that can be easily shared across different deep learning frameworks.
- 1-bit Stochastic Gradient Descent There is no separate license required to use the 1-bit Stochastic Gradient Descent (1-bit SGD) in CNTK. This makes it a useful tool for those who want to use this specific feature without having to pay for a separate license.
Reviews
0 reviews
Leave a review