A distributed platform for rapid deep learning application development with neural network engine and Hadoop integration.
Veles is a distributed machine learning platform developed by Samsung for rapid deep learning application development. It consists of a core platform along with specialized plugins for neural network training, Hadoop integration, and audio feature extraction. The platform enables researchers and engineers to build and deploy machine learning models across distributed systems efficiently.
Machine learning researchers, data scientists, and engineers working on deep learning projects that require distributed computing capabilities and integration with big data ecosystems like Hadoop.
Developers choose Veles for its integrated approach to distributed deep learning, offering specialized components for common tasks while maintaining compatibility with existing big data tools through its Java bridge, reducing development complexity.
Distributed machine learning platform
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Provides a comprehensive framework for developing deep learning applications across multiple nodes, as highlighted in the key features for distributed systems.
Znicz plugin is dedicated to building and training neural networks, offering optimized performance for model development.
Mastodon bridge enables seamless integration with Hadoop and other Java tools, facilitating big data workflows as described in the README.
SoundFeatureExtraction library provides specialized tools for audio feature extraction, useful for audio analysis tasks mentioned in the features.
Last update was in 2015, which may not support recent deep learning advancements, modern hardware, or current best practices.
As a Samsung project from 2013-2015, it likely has scarce community contributions and outdated documentation compared to active projects like TensorFlow or PyTorch.
Requires integration with Hadoop and multiple nodes, which can be challenging for teams without prior experience in distributed systems, adding overhead.