Summary of Onnxpruner: Onnx-based General Model Pruning Adapter, by Dongdong Ren et al.
ONNXPruner: ONNX-Based General Model Pruning Adapter
by Dongdong Ren, Wenbin Li, Tianyu Ding, Lei Wang, Qi Fan, Jing Huo, Hongbing Pan, Yang Gao
First submitted to arxiv on: 10 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes ONNXPruner, a novel pruning adapter designed for ONNX format models, which streamlines the adaptation process across diverse deep learning frameworks and hardware platforms. The adapter uses node association trees to automatically adapt to various model architectures, clarifying structural relationships between nodes and guiding the pruning process. Additionally, the authors introduce a tree-level evaluation method that enhances pruning performance without requiring extra operations. Experiments confirm ONNXPruner’s strong adaptability and increased efficacy across multiple models and datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps make it easier to remove parts of deep learning models without affecting their performance. The authors created a special tool called ONNXPruner that can be used with different types of deep learning frameworks and hardware platforms. This tool uses a special kind of tree structure to understand how the model’s nodes are connected, which helps guide the pruning process. The authors also came up with a new way to measure how well the pruning works, which takes into account how all the nodes in the model work together. |
Keywords
» Artificial intelligence » Deep learning » Pruning
