About
In Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer, we show that optimal hyperparameters become stable across neural network sizes when we parametrize the model in maximal update parametrization (μP). This can be used to tune extremely large neural networks such as large pretrained transformers, as we have done in our work. More generally, μP reduces the fragility and uncertainty when transitioning from exploration to scaling up, which are not often talked about explicitly in the deep learning literature.
Open Source Health
- Stars
- 1,755
- Forks
- 105
- License
- MIT
- Last commit
- 2 years ago
Resources & Links
Related Categories
Vendor
Microsoft
Empower every person and organization to achieve more.
Quick Links
Open Source
More by Microsoft
Related Products
Rclip
Semantic photo search for the command line
Top category match
ChineseNER
A neural network model for Chinese named entity recognition
Top category match
Head Pose Estimation
Realtime human head pose estimation with ONNXRuntime and OpenCV.
Top category match
Video Subtitle Extractor
视频硬字幕提取，生成srt文件。无需申请第三方API，本地实现文本识别。基于深度学习的视频字幕提取框架，包含字幕区域检测、字幕内容提取。A GUI tool for extracting hard-coded subtitle (hardsub) from videos and generating srt files.
Top category match
U 2 Net
The code for our newly accepted paper in Pattern Recognition 2020: "U^2-Net: Going Deeper with Nested U-Structure for Salient Object Detection."
Top category match