Cognitive Toolkit: Open Source Deep Learning SW

Cognitive Toolkit: Open Source Deep Learning SW

Cognitive Toolkit: Microsoft Open Source Deep Learning SW

Microsoft's Cognitive Toolkit (formerly called CNTK) is a deep learning toolkit (Machine Learning) de «Código Abierto» with enormous potential. It's also free, easy to use, and has a commercial grade quality that allows to create deep learning algorithms capable of learning at a level close to that of the human brain.

Microsoft, its creator, ensures that this open source tool provides uncompromising scaling, good quality speed and precision, and compatibility with widely used programming languages ​​and algorithms today.

Cognitive Toolkit: Introduction

And that, consequently, is capable of training and evaluating deep learning algorithms faster than other similar tools. While, in turn, it is capable of scaling efficiently in a wide range of environments (CPUs, GPUs, even Cluster of Equipment) without losing precision.

It built with sophisticated algorithms and production readers to work reliably with massive data sets. The same is applied in products such as Skype, Cortana, Bing and Xbox. And it is also currently used in the great technology industry to develop «Inteligencia Artificial (IA)» commercial grade. Above all, for its support for C ++ and Python, which allows you to customize almost any existing or your own training algorithm.

Cognitive Toolkit: Content

Cognitive Toolkit: What is it and what are its characteristics?

  • It is an open source toolkit for deep learning.
  • It is a product developed by Microsoft whose latest stable version is 2.7.
  • Describe neural networks as a series of computational steps through a directed graph.
  • It allows users to easily make and combine the most popular types of models, such as: Deep Neural Networks, Convolutional and Recurrent.
  • Implement stochastic gradient descent learning with automatic differentiation and parallelization across multiple GPUs and servers.
  • It can be included as a library in Python, C # or C ++ programs, or used as a standalone deep learning tool through its own model description language (BrainScript). Or make use of its evaluation functionality from programs created with Java.
  • It is compatible with Linux 64-bit or Windows 64-bit Operating Systems through precompiled binary packages, or through compilation using their source files hosted on GitHub.

Final report: For more information about the Cognitive Toolkit you can visit their Official website at Microsoft or its website at GitHub.


El «Intercambio de Redes Neuronales Abiertas (Open Neural Network Exchange - ONNX)» is an open ecosystem that allows developers of «IA» choose the right tools as your project evolves.

Y Cognitive Toolkit is one of the first deep learning toolkits to support the format «ONNX». Format what is co-developed by Microsoft and supported by many others. So that, «ONNX» allows developers to move models between frameworks like Cognitive Toolkit, Caffe2, MXNet, and PyTorch.

Finally, «ONNX» provides an open source format for models «IA» and define a graphical extensible computation modelas well as includes definitions of built-in operators and standard data types. And it focuses on the capabilities needed for inference processes (scoring).

Final report: For more information on «ONNX» You can visit the official website of the same in the following link Open Neural Network Exchange.

Cognitive Toolkit: Conclusion


We hope you are "small but useful post" about this other interesting software «Código Abierto» of Microsoft Open Application Ecosystem calls «Cognitive Toolkit», is of great interest and utility, for the entire «Comunidad de Software Libre y Código Abierto» and of great contribution to the diffusion of the wonderful, gigantic and growing ecosystem of applications of and for «GNU/Linux».

And for more information, always do not hesitate to visit any Online library Be OpenLibra y jedit to read books (PDFs) on this topic or others knowledge areas. For now, if you liked this «publicación», don't stop sharing it with others, in your Favorite websites, channels, groups, or communities of social networks, preferably free and open as Mastodon, or secure and private like Telegram.

Or simply visit our home page at FromLinux or join the official Channel Telegram from DesdeLinux to read and vote for this or other interesting publications on «Software Libre», «Código Abierto», «GNU/Linux» and other topics related to «Informática y la Computación», And the «Actualidad tecnológica».

The content of the article adheres to our principles of editorial ethics. To report an error click here!.

Be the first to comment

Leave a Comment

Your email address will not be published. Required fields are marked with *



  1. Responsible for the data: Miguel Ángel Gatón
  2. Purpose of the data: Control SPAM, comment management.
  3. Legitimation: Your consent
  4. Communication of the data: The data will not be communicated to third parties except by legal obligation.
  5. Data storage: Database hosted by Occentus Networks (EU)
  6. Rights: At any time you can limit, recover and delete your information.