SonySemiconductorSolutions/mct-model-optimization
A quantization and compression toolkit for optimizing neural network inference on efficient, constrained hardware targets.
Not currently ranked — collecting fresh signals.
star history
Model Compression Toolkit (MCT) provides advanced post-training quantization (PTQ) and quantization-aware training (QAT) capabilities for neural networks. It supports both PyTorch and TensorFlow frameworks, enabling researchers and engineers to compress models for edge deployment while maintaining accuracy. The toolkit targets efficient inference on hardware-constrained environments.
Frequently asked
- What is SonySemiconductorSolutions/mct-model-optimization?
- A quantization and compression toolkit for optimizing neural network inference on efficient, constrained hardware targets.
- Is mct-model-optimization open source?
- Yes — SonySemiconductorSolutions/mct-model-optimization is open source, released under the Apache-2.0 license.
- What language is mct-model-optimization written in?
- SonySemiconductorSolutions/mct-model-optimization is primarily written in Python.
- How popular is mct-model-optimization?
- SonySemiconductorSolutions/mct-model-optimization has 448 stars on GitHub.
- Where can I find mct-model-optimization?
- SonySemiconductorSolutions/mct-model-optimization is on GitHub at https://github.com/SonySemiconductorSolutions/mct-model-optimization.