NervanaSystems/neon
Intel's reference deep learning framework for training neural networks with optimized performance on CPUs and GPUs.
Not currently ranked — collecting fresh signals.
star history
Neon is Intel’s deep learning framework designed for neural network training and experimentation. It provides implementations of common deep learning layers including convolution, RNN, LSTM, GRU, and BatchNorm, along with a model zoo containing pre-trained weights for image classification, reinforcement learning, and NLP tasks. The framework offers swappable hardware backends supporting CPUs and GPUs for deployment flexibility.
Frequently asked
- What is NervanaSystems/neon?
- Intel's reference deep learning framework for training neural networks with optimized performance on CPUs and GPUs.
- Is neon open source?
- Yes — NervanaSystems/neon is open source, released under the Apache-2.0 license.
- What language is neon written in?
- NervanaSystems/neon is primarily written in Python.
- How popular is neon?
- NervanaSystems/neon has 3.9k stars on GitHub.
- Where can I find neon?
- NervanaSystems/neon is on GitHub at https://github.com/NervanaSystems/neon.