Open Baxulio opened 11 months ago
You replaced the neural network files with files from someone else, trained in a different way, perhaps not even Darknet/YOLO files, and now you'd like me to debug things for you?
Stick with the neural network files I provide, or train your own as described in this video: https://www.youtube.com/watch?v=jz97_-PCxl4
Good Morning Dear Mr. Stéphane Charette,
In this published paper http://sergiomsilva.com/pubs/alpr-unconstrained/ authors already provided well-trained models, including Darknet trained network. One can get them by this script:
wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.cfg -P data/ocr/
wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.names -P data/ocr/
wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.weights -P data/ocr/
wget -c -N http://sergiomsilva.com/data/eccv2018/ocr/ocr-net.data -P data/ocr/
I just wondered whether it's possible to adopt these files in DarkPlate without retraining or anything else. Now I understand that this way doesn't work and DarkPlate has its own structure of files.
Anyway thank you for the great project and your answer!
I have no idea, you'd have to try it out.
Hello Dear Mr. Stéphane Charette,
I tried the DarkPlate executable app with the config files and weghts you provided and it works! But if I replace them with the files and weights from this url (http://sergiomsilva.com/pubs/alpr-unconstrained/) it doesn't work.
App exits here:
How can I adopt aforementioned pretrained models? Thanks in advance!