https://github.com/noopur-zambare/nc_pruning
Dense network pruning algorithm inspired from Neural Collapse geometry
https://github.com/noopur-zambare/nc_pruning
deep-learning robust-machine-learning statistics trustworthy-ai
Last synced: about 2 months ago
JSON representation
Dense network pruning algorithm inspired from Neural Collapse geometry
- Host: GitHub
- URL: https://github.com/noopur-zambare/nc_pruning
- Owner: noopur-zambare
- Created: 2024-12-08T02:51:26.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-12-23T17:39:08.000Z (over 1 year ago)
- Last Synced: 2025-03-01T15:14:07.475Z (over 1 year ago)
- Topics: deep-learning, robust-machine-learning, statistics, trustworthy-ai
- Language: Jupyter Notebook
- Homepage:
- Size: 12.2 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
## Dense Network Pruning using Neural Collapse under Imbalanced Dataset
#### Install the required dependencies
```
pip install -r requirements.txt
```
#### To run the code
```
python3 main.py
```
We have also provided a ipynb file for easy execution.
## Results
Testing Pruned Model
Testing Pruned Model with Noisy Data
```
Folder Structure
├── logs
│ ├── testing_with_noise
│ ├── testing_without_noise
├── results
│ ├── testing_with_noise
│ ├── testing_without_noise
├── main.py
├── pruning.py
├── sensitivity_analysis.py
├── README.md
├── requirements.txt
└── main.ipynb
```