Hello:
I need to train ocropus, y will comment the steps that execute for
this, i need to know if this steps are ok or wrong:
this is the structure of the directories that i use:
/root
/root/list.txt
/root/training
/root/training/01.png
/root/training/01.txt
/root/training/02.png
/root/training/02.txt
1. I take one page, and cut 2 lines, convert the image to B/W with png
format without compresion. i call that images lines 01.png and 02.png
2. I create two txt, one for each image line, i called:
a) 01.txt: content the words of the lines in the image 01.png,
for example:
t
h
e
u
g
l
y
b) 02.txt: same as a), but with thw words of the image line
called 02.png:
T
E
C
N
O
L
O
G
Y
3. move 01.png, 02.png, 01.txt, 02.txt to the folder "training"
4. create the "list.txt", that content:
./training/02.png ./training/02.txt
./training/01.png ./training/01.txt
5. ejecute ocroscript:
r...@us810:~/# ocroscript align-lines-wordwise ./list.txt /usr/local/
share/ocropus/models/neural-net-file.nn
././0/02.txt
[FAILURE] t
h
e
u
g
l
y
.
### BAD LINE: max cost in word `T
E
C
N
O
L
O
G
Y
.' is 199999993605713849301312521538346418176.000000
The step 5 create two files: 01.cost and 02.cost, but are empty....
What im are doing wrong? that are the correct steps?
Best regards.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"ocropus" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/ocropus?hl=en
-~----------~----~----~----~------~----~------~--~---
