https://github.com/jason-cs18/chameleon
Chameleon: An efficient continuous adaptation framework based on NVIDIA TAO.
https://github.com/jason-cs18/chameleon
continuous-learning domain-adaptation model-selection nvidia-tao-toolkit
Last synced: 2 months ago
JSON representation
Chameleon: An efficient continuous adaptation framework based on NVIDIA TAO.
- Host: GitHub
- URL: https://github.com/jason-cs18/chameleon
- Owner: Jason-cs18
- Created: 2021-11-29T13:48:19.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2021-11-29T14:21:18.000Z (over 4 years ago)
- Last Synced: 2024-05-18T17:58:45.957Z (about 2 years ago)
- Topics: continuous-learning, domain-adaptation, model-selection, nvidia-tao-toolkit
- Homepage:
- Size: 1.95 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Chameleon
Chameleon is an efficient continuous adaptation framework based on [NVIDIA TAO](https://developer.nvidia.com/zh-cn/tao-toolkit-get-started). To bridge the gap between one-time domain adaptation and continuous learning, we propose Chameleon, which updates models on new data (labeled or unlabeled) via **existing domain adaptation techniques** and select the suitable model to recovery accuracy through **adaptive model selection methods**. In implementation, we provide different adaptation strategies and optimization techniques for different visual tasks (object detection, segmentation, tracking and SLAM). In the end, we summary existing common optimization techniques (GPU sharing, ...).
## 1. Installation (TAO and Docker)
## 2. Scenarios (adaptation strategies and optimization techniques)
### Object Detection
### Image Segmentation
### Visual Tracking
### SLAM (in progress)
## 3. Common optimization techniques
- GPU Sharing between training and inference: [Ekya: Continuous Learning of Video Analytics Models on Edge Compute Servers. Published in _NSDI'22_.](https://www.microsoft.com/en-us/research/publication/ekya-continuous-learning-of-video-analytics-models-on-edge-compute-servers-2/)