Skip to main content
SpringerPlus

Table 4 Error rates for Urdu Nasta’liq text line recognition for training and validation sets

From: Urdu Nasta’liq text recognition using implicit segmentation based on multi-dimensional long short term memory neural networks

Errors

Training set (%)

Validating set (%)

Label error

0.85

1.83

Deletions

0.22

0.48

Insertions

0.150

0.22

Substitutions

0.480

1.21

CTC error

3.11

8.28

Back to article page