Tools

The tools/ folder contains helper scripts and utilities that support data conversion, visualization, HPC execution, and optimization workflows in VISWIR.

Warning

The scripts and utilities provided in the tools/ folder are distributed as-is, without any warranty. Their use is entirely at the responsibility of the user.

The authors cannot be held liable for any data loss, corruption, or other issues that may arise from their use. Please ensure you test scripts on non-critical data before applying them in production environments.

Data Conversion

  • coco_to_voc_converter.py : convert COCO annotations to Pascal VOC format.

  • export_data_to_csv.py : export results or metrics to CSV.

  • export_data_to_json.py : export results or metrics to JSON (optional).

Visualization

  • draw_graph_from_csv.py : generate graphs from CSV outputs (To be used after using the export_data_to_csv.py script).

  • draw_annotations_on_images.py : overlay annotations on images (optional).

  • csv_graph_comparator.py : compare multiple CSV result files (optional).

HPC / Container

  • Singularity : Singularity container definition file.

  • job_viswir.slurm : SLURM job submission script for HPC environments.

  • job_viswir_ex.slurm : extended SLURM job script (custom resources and option).

  • job_viswir_test.slurm : lightweight SLURM script for testing.

Optimization

  • optuna_sqlite3_loader_mono_obj.py : load and analyze Optuna results (single objective - F1).

  • optuna_sqlite3_loader_multi_obj.py : load and analyze Optuna results (multi-objective - NR-IQA).

  • sql_explorer.py : explore the SQLite results database interactively.

Usage

Each script is standalone and must be adapted to your environment. Refer to the script headers for usage instructions.

Tip

  • HPC users should start with job_viswir.slurm and the Singularity file.

  • For optimization analysis, use the Optuna loaders together with sql_explorer.py.

  • Visualization scripts are mainly useful for legacy or debugging workflows.