rodrigetmirts_submission_2026

bohdanUWU/rodrigetmirts_submission_2026

rodrigetmirts_submission_2026 is a machine learning model from bohdanUWU. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

These are alternative script.py versions that scored worse than the current best (0.4584) but did NOT completely fail. Kept here in case you want to re-deploy them, A/B test variants, or build on them.

Downloads · 30 days

Access

Public

Updated May 8, 2026

Repo size

—

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.py97.3 KB · 95%

At a glance

Access: Public
Created: May 8, 2026
Updated: May 8, 2026
SHA: 849dc973

Model card

From the Hugging Face model README

Folder 1 — backup of worse-but-not-bad methods

These are alternative script.py versions that scored worse than the current best (0.4584) but did NOT completely fail. Kept here in case you want to re-deploy them, A/B test variants, or build on them.

This folder is outside the HuggingFace repo (s23-model/) so files here will NEVER get pushed to the leaderboard accidentally.

Files

File	HF commit	Leaderboard score	Notes
`script_colmap_junction.py` + `postprocess_v2_colmap_junction.py`	`6cf3fbd`	hss=0.4443, corner=0.5003, edge=0.4101	COLMAP multi-view plane refinement + junction constraints + snap class 3 + safer snap_horizontal + loosened hybrid_merge. All bundled. -0.014 vs baseline.
`script_pipeline_ensemble.py`	`3118106`	hss=0.4406, corner=0.4856, edge=0.4122	Learned + classical (sklearn_submission) pipeline ensemble with heavy features disabled to avoid timeout. -0.018 vs baseline. The "lite ensemble" variant.
`script_ensemble_full_TIMEOUT.py`	`9cbefb9`	TIMEOUT (0.0)	Full classical-pipeline ensemble — predict_wireframe_sklearn called with all default features ON (USE_TRACK_ENSEMBLE etc). Re-runs DLT triangulation that hybrid_merge already does. Hit the 2h budget. Worst possible outcome (all zeros).
`script_tta_unoptimized_TIMEOUT.py`	`8e33a89`	TIMEOUT (0.0)	First TTA attempt with K=3 passes, but each pass re-ran build_compact_scene (the expensive multi-view label voting). 3x the slowest step → timeout. The fix is now in production: cache fusion, vary only priority sampling.
`script_degraded_~0.32.py`	(synthetic, not on HF)	~0.30–0.35 estimated	Deliberately degraded variant. Starts from the 0.4584 baseline (commit 56f1ec6) and: (1) raises CONF_THRESH 0.4→0.7 (drops mid-confidence segments), (2) disables snap_to_point_cloud + snap_horizontal (raw model coords without semantic refinement), (3) disables classical track ensemble (no predict_wireframe_tracks + hybrid_merge). The model's overconfident skeleton without any post-processing.

Why kept

Even though both regressed, they may still be useful:

script_colmap_junction.py — the individual fixes inside (snap class 3, safer snap_horizontal) might help in isolation; the bundle is what hurt. The COLMAP refine code is fully integrated and could be tuned (try min_quality=0.7, max_displacement=0.2 instead of the loose defaults that snapped vertices to wall planes).
script_pipeline_ensemble.py — the predict_wireframe_sklearn plumbing is wired up. If you ever want to do a SELECTIVE ensemble (e.g., only use classical-pipeline output for samples where learned model has low confidence), this is the starting point.

To deploy any of these

# Example: deploy the COLMAP-junction variant to the HF repo
cp 1/script_colmap_junction.py s23-model/script.py
cp 1/postprocess_v2_colmap_junction.py s23-model/s23dr_2026_example/postprocess_v2.py
cd s23-model && git add . && git commit -m "Try variant X" && git push

Current best (in s23-model/, HEAD)

script.py at commit 857514e: TTA optimized (3-pass with cached fusion). Awaiting leaderboard score.

Previous bests (in s23-model git history)

Commit	Score	Description
`9deb9e1`	hss=0.4583	Hybrid geometric ensemble (pre-experiment baseline)
`61a6827`	hss=0.4583	Pure revert to 9deb9e1
`56f1ec6`	hss=0.4584	CONF_THRESH 0.5→0.4 (current production)

Created: May 8, 2026
Updated: May 8, 2026