trocr-base-ru / README.md
raxtemur's picture
Update README.md
f981e8b verified
|
raw
history blame
757 Bytes
metadata
license: apache-2.0
datasets:
  - nastyboget/stackmix_hkr_large
  - nastyboget/stackmix_cyrillic_large
  - nastyboget/synthetic_cyrillic_large
language:
  - ru
  - en
pipeline_tag: image-to-text
tags:
  - ocr

Model Card for TrOCR-Ru

Finetuned model microsoft/trocr-base-handwritten on large synth datasets from nastyboget.

Metrics on HKR/Cyrillic datasets

Metric HKR_test1 HKR_test2 CYR_test
Accuracy 60.71 62.48 58.29
CER 10.47 8.33 10.30
WER 34.10 27.86 38.53