Text recognition from an images
create dataset with structure:
dataset
|
...train
...|
......classA
......|
.........imgA
.........imgB
............
.........imgZ
......classB
...
...validation
...|
......classA
......classB
...
Set CLASSES in ./main.py
Change Dense
with class count in ./train.py
near model = Sequential([