Codesota · Computer Vision · Optical Character Recognition · pendigitsTasks/Computer Vision/Optical Character Recognition
Optical Character Recognition · benchmark dataset · 2020 · EN

pendigits.

Dataset from Papers With Code

Saturated benchmark

Benchmark near ceiling or stagnant — no meaningful SOTA movement in 2+ years

Submit a result ↵
§ 01 · Leaderboard

Best published scores.

20 results indexed across 4 metrics. Shaded row marks current SOTA; ties broken by submission date.

Primary
accuracy · higher is better
All metrics
accuracy, ari, nmi, runtime-s
accuracy· primary
8 rows
#ModelOrgSubmittedPaper / codeaccuracy
01DnC-SC—May 2021papers-with-code · code82.27
02U-SPEC—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code81.68
03LSC-R—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code81.55
04ASNMF-SRPZhong and GaoAug 2025source80.44
05RCCAcademicJul 2024Deep Online Probability Aggregation Clustering79.60
06PACYan et al.Jul 2024Deep Online Probability Aggregation Clustering78
07LBDM—Jun 2018papers-with-code74.70
08LSC-K—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code74.02
ari
1 row
#ModelOrgSubmittedPaper / codeari
01ASNMF-SRPZhong and GaoAug 2025source68.49
nmi
5 rows
#ModelOrgSubmittedPaper / codenmi
01DnC-SC—May 2021papers-with-code · code82.86
02U-SPEC—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code81.68
03LSC-K—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code81.37
04ASNMF-SRPZhong and GaoAug 2025source80.12
05LSC-R—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code79.15
runtime-s
6 rows
#ModelOrgSubmittedPaper / coderuntime-s
01LBDM—Jun 2018papers-with-code3.08
02U-SPEC—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code2.07
03SC_RB—May 2018Scalable Spectral Clustering Using Random Binning Featur… · code1.80
04LSC-K—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code1.20
05LSC-R—Apr 2021Divide-and-conquer based Large-Scale Spectral Clustering · code0.770
06DnC-SC—May 2021papers-with-code · code0.640
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 03 · Progress

3 steps
of state of the art.

Each row below marks a model that broke the previous record on accuracy. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Higher scores win. Each subsequent entry improved upon the previous best.

SOTA line · accuracy
  1. Jun 1, 2018LBDM74.70
  2. Apr 30, 2021U-SPEC81.68
  3. May 2, 2021DnC-SC82.27
Fig 3 · SOTA-setting models only. 3 entries span Jun 2018 → May 2021.
§ 04 · Literature

3 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies
pendigits — Optical Character Recognition benchmark · Codesota | CodeSOTA