search

qqwweee / keras-yolo3

A Keras implementation of YOLOv3 (Tensorflow backend)
MIT License
7.14k stars 3.45k forks source link

Load custom trained model problem #498

Open houzeyu2683 opened 5 years ago

houzeyu2683 commented 5 years ago

I want to write a detect function for my model, It look like detect(image) and return class_prob, bounded box. I use tiny-yolo model. I already train and save one model(use model.save not model.save_weights) in the logs/000/ folder.

But I get error when I load model. The first error message is NameError: name 'yolo_head' is not defined So I import yolo_head. The second error message is TypeError: list indices must be integers or slices, not list

Is there any better idea to get the bounded box from image when finish training own model? Thanks

Borda commented 5 years ago

pls, have look at https://github.com/Borda/keras-yolo3/issues/2