Spaces:
Sleeping
Sleeping
Commit History
-
Update training configuration and dependencies
- Modified `requirements.txt` to use flexible versions of PyTorch and related libraries. - Adjusted training parameters in `train.yaml`, reducing epochs from 20 to 8 and batch sizes from 32 and 64 to 8. - Enhanced `train_gnn.py` with gradient accumulation and mixed precision training support. - Fixed data type handling in `graph_models.py` for message passing layer.1508177
-
Add Gradio app for inference and model selection
- Introduced app.py for a Gradio interface allowing users to upload images and select between baseline and GNN models for inference. - Implemented ModelRegistry for caching loaded models and managing checkpoints. - Added README_SPACE.md to guide users on setting up the Hugging Face Space with the necessary files and configurations. - Created requirements-space.txt for dependencies specific to the Gradio app. - Developed scripts for single-image inference and comparison of baseline vs GNN model metrics. - Enhanced existing README.md with updated training and evaluation instructions for both models.419fc11 -
Add data pipeline scripts and configuration files for Visual Genome processing
Ashish Mehta- Introduced .gitignore to exclude unnecessary files and directories. - Added requirements.txt for project dependencies. - Implemented dataset.yaml with download, extraction, and processing configurations. - Created multiple scripts for data handling: download, extract, process, and build splits. - Developed run_baselines.py and eval_baseline.py for baseline model evaluation. - Enhanced README.md with data pipeline instructions and expected outputs. - Established src/data module for data utilities and processing functions.4687353