Master IASD · DataLab 2026

Collaborative filtering

A1 · Daily results

Updated 01 October 2026 · 05:05 CESTDaily run around 05:00 · Paris time

Podium

Lowest RMSE wins. Student groups and professor references compete together.

1
Professor reference

BestOf 2023 · 1

0.803 RMSE

2
Professor reference

BestOf 2025 · 1

0.818 RMSE

3
Professor reference

BestOf 2025 · 2

0.830 RMSE

Daily progress

RMSE over time · lower is better. Dots are daily evaluations, not GitHub activity. Failed runs leave gaps; no submission means no point.

Results

Scores use hidden evaluation ratings; the input contains train + test. Accuracy is the percentage of exactly predicted ratings. These measurements support your analysis; code itself is not graded.

Hover or focus the ! icon to read an error. Click it to copy the message.

RankStatusSource
1
Professor reference
BestOf 2023 · 1Reference
0.8030.0041.87Success Frozen reference8b6fb772
2
Professor reference
BestOf 2025 · 1Reference
0.8180.0098.34Success Frozen referenceab53d900
3
Professor reference
BestOf 2025 · 2Reference
0.8300.024.78Success Frozen reference55d5e886
4
Professor reference
BestOf 2024 · 1Reference
0.8470.00149.52Success Frozen reference1204cccc
5
Professor reference
BestOf 2023 · 2Reference
0.86026.1198.94Success Frozen reference2fa4d93a
6
Professor reference
BestOf 2024 · 2Reference
0.86625.3142.55Success Frozen reference7869cd93
7
Professor reference
averageReference
1.0370.001.18Success Frozen reference
8
Professor reference
randomReference
1.82314.281.18Success Frozen reference
BestOf references are preserved benchmarks from previous years. Their current scores are recomputed with this platform; archived scores remain in the history.

Run 20261001T030001Z · Execution limit: 1200 s per group. Student runs: 4 CPU threads · 12 GiB host memory · CPU only. Historical references use one A40 GPU when required. Dependency installation time is not included in the reported execution time.