Open raghavgurbaxani opened 4 years ago
There is experimental support for an (untrained) EfficientNet backbone which you are able to train with the package. You can use keras_ocr.detection.Detector(weights=None, backbone_name='EfficientNetB0')
.
Hi, thanks for your work! One quick question, you mentioned in issue #36 that you were experimenting with efficientnet. Did you manage to train the network and get good performance? If that's the case, it would be great if you share with us your result! Thanks
Hi
Thanks for the great work on this. Is it possible to retrain CRAFT based on a lighter MobileNet or EfficientNet backbone ? Thanks...