https://github.com/cserajdeep/liquid-neural-network-lnn
LIQUID NEURAL NETWORK LNN CLASSIFIER AND REGRESSION
https://github.com/cserajdeep/liquid-neural-network-lnn
boston-housing-dataset boston-housing-price-prediction classification classifier cse-rajdeep huggingface iris-dataset kaggle-dataset kiit liquid-neural-networks lnn-classifier open-source regression
Last synced: 27 days ago
JSON representation
LIQUID NEURAL NETWORK LNN CLASSIFIER AND REGRESSION
- Host: GitHub
- URL: https://github.com/cserajdeep/liquid-neural-network-lnn
- Owner: cserajdeep
- License: mit
- Created: 2025-03-14T06:46:01.000Z (about 1 month ago)
- Default Branch: main
- Last Pushed: 2025-03-14T06:49:29.000Z (about 1 month ago)
- Last Synced: 2025-03-14T07:32:25.629Z (about 1 month ago)
- Topics: boston-housing-dataset, boston-housing-price-prediction, classification, classifier, cse-rajdeep, huggingface, iris-dataset, kaggle-dataset, kiit, liquid-neural-networks, lnn-classifier, open-source, regression
- Language: Jupyter Notebook
- Homepage:
- Size: 0 Bytes
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🧠Run-1 LIQUID NEURAL NETWORK LNN🚀
## True Liquid Time-Constant (LTC) Cell
- Implemented proper continuous-time dynamics with learnable time constants
- Added decay factor based on tau parameter that controls information flow
---
## Enhanced Architecture
- Multi-layer capability for deeper networks
- Self-attention mechanism for capturing temporal relationships
- Skip connections and layer normalization for better gradient flow
---
## Robust Training Framework
- Learning rate scheduling with `ReduceLROnPlateau`
- Early stopping to prevent overfitting
- Gradient clipping to prevent exploding gradients
- AdamW optimizer with weight decay for regularization
---
## Comprehensive Evaluation
- Detailed metrics for both classification and regression
- Visualization of results (confusion matrices, prediction plots)
- Feature importance analysis for regression tasks
---
## Hyperparameter Tuning
- Simple grid search to find optimal model configuration
- Best model checkpointing
---
## Improved Data Handling
- Better error handling for dataset loading
- Proper input normalization
- Support for both sequence and non-sequence data formats
---
The code is now much more robust, handles edge cases better, and should provide significantly better performance on both classification and regression tasks.