Optical Character Recognition2020en

codesearchnet---python

Dataset from Papers With Code

Metrics:accuracy, cer, wer, f1

smoothed-bleu-4

#ModelScorePaper / CodeDate
1
CodeTrans-MT-Base
20.39
CodeTrans: Towards Cracking the Language of Silicon's Code Through Self-Supervised Deep Learning and High Performance ComputingCode
Apr 2021
2
CodeBERT (MLM)
15.48
CodeBERT: A Pre-Trained Model for Programming and Natural LanguagesCode
Feb 2020
3
CodeBERT (MLM+RTD)
15.41
CodeBERT: A Pre-Trained Model for Programming and Natural LanguagesCode
Feb 2020
4
pre-train w/ code only
15.12
CodeBERT: A Pre-Trained Model for Programming and Natural LanguagesCode
Feb 2020
5
RoBERTa
14.92
CodeBERT: A Pre-Trained Model for Programming and Natural LanguagesCode
Feb 2020
6
Transformer
13.44
CodeBERT: A Pre-Trained Model for Programming and Natural LanguagesCode
Feb 2020
7
seq2seq
13.04
CodeBERT: A Pre-Trained Model for Programming and Natural LanguagesCode
Feb 2020

Related Papers2

CodeTrans: Towards Cracking the Language of Silicon's Code Through Self-Supervised Deep Learning and High Performance Computing
Apr 2021Models: CodeTrans-MT-Base
PaperCode
CodeBERT: A Pre-Trained Model for Programming and Natural Languages
Feb 2020Models: CodeBERT (MLM), CodeBERT (MLM+RTD), pre-train w/ code only +3 more
PaperCode

Other Optical Character Recognition Datasets

scut-ctw1500mpqacnn-/-daily-mailicdar2013SROIEKITAB-BenchThaiOCRBenchPolEval 2021 OCRIMPACT-PSNCwikibio
codesearchnet---python Benchmark - Optical Character Recognition | CodeSOTA