Turn training results into structured experiment comparisons, ablation analysis, Pareto evaluation, repeated-run statistics, and publication-ready tables and figures.
Designed around common computer vision outputs including Ultralytics, MMDetection, MMSegmentation, Detectron2, TensorBoard, training logs and CSV results.
Build comparison tables, identify best and second-best results, check experimental settings, and distinguish validation from test-set evidence.
Analyse component contributions, missing ablation combinations, efficiency trade-offs, Pareto-efficient models, and repeated runs.
Prepare LaTeX, Word and Excel tables plus journal-style vector and high-resolution figures for research reporting.
ExperimentOS is currently in testing and is not yet publicly available. Join the notification list to hear when access opens.