haitongli/knowledge-distillation-pytorch
A PyTorch framework for knowledge distillation experiments on CIFAR-10 using teacher-student model training.
Not currently ranked — collecting fresh signals.
star history
This repository provides a PyTorch implementation for exploring knowledge distillation (KD) experiments on CIFAR-10. It implements a teacher-student training paradigm where a smaller student network learns from a larger pretrained teacher model (e.g., ResNet-18) using dark knowledge. The framework includes hyperparameter search utilities, tensorboard support, checkpoint management, and pretrained teacher models for download.
Frequently asked
- What is haitongli/knowledge-distillation-pytorch?
- A PyTorch framework for knowledge distillation experiments on CIFAR-10 using teacher-student model training.
- Is knowledge-distillation-pytorch open source?
- Yes — haitongli/knowledge-distillation-pytorch is open source, released under the MIT license.
- What language is knowledge-distillation-pytorch written in?
- haitongli/knowledge-distillation-pytorch is primarily written in Python.
- How popular is knowledge-distillation-pytorch?
- haitongli/knowledge-distillation-pytorch has 2k stars on GitHub.
- Where can I find knowledge-distillation-pytorch?
- haitongli/knowledge-distillation-pytorch is on GitHub at https://github.com/haitongli/knowledge-distillation-pytorch.