starride-teklia
commited on
Commit
•
6f2b942
1
Parent(s):
2b89df8
Update README.md
Browse files
README.md
CHANGED
@@ -30,7 +30,7 @@ For training, text-lines were resized with a fixed height of 128 pixels, keeping
|
|
30 |
| val | 3,102 |
|
31 |
| test | 3,819 |
|
32 |
|
33 |
-
An external 6-gram character language model can be used to improve recognition. The language model is trained on the text from the
|
34 |
|
35 |
## Evaluation results
|
36 |
|
|
|
30 |
| val | 3,102 |
|
31 |
| test | 3,819 |
|
32 |
|
33 |
+
An external 6-gram character language model can be used to improve recognition. The language model is trained on the text from the Belfort training set.
|
34 |
|
35 |
## Evaluation results
|
36 |
|