- Implemented `tracing()` method in training pipeline to export model as TorchScript
- Created dedicated TorchScript directory in checkpoints path
- Added model tracing with example input tensor
- Saved traced model as `model_h.pt` in the TorchScript directory
- Updated Docker run command with improved GPU and user settings
- Modified learning rate scheduler to support epoch-based adjustments
- Refactored loss computation to return detailed loss components
- Added TensorBoard logging for individual loss components
- Implemented checkpoint saving mechanism with configurable frequency
- Updated training script to use dynamic configuration and improved error handling
- Added `position_loss` method to calculate manifold loss using mean absolute value
- Implemented `normals_loss` method to compute normal vector loss with gradient calculation
- Updated `train.py` to log input and output tensor shapes during training
- Modified network architecture in `train.py` by increasing hidden layer dimensions from [64, 64, 64] to [256, 256, 256]
- Created `loss.py` with a `LossManager` class to handle loss calculation
- Integrated `LossManager` into the training pipeline in `train.py`
- Implemented a basic manifold loss computation using mean absolute value
- Added `data_loader.py` with `NHREP_Dataset` class for loading point cloud, feature mask, and CSG tree data
- Implemented `CustomDataLoader` for flexible data loading with configurable parameters
- Refactored `train.py` to create a structured training pipeline for NHRepNet
- Added support for feature sampling, device selection, and TensorBoard logging
- Introduced modular training methods with error handling and logging
mckay