TypeError: type object got multiple values for keyword argument 'training'

Viewed 2028

I am checking an object detection collaboratory notebook and it's supposed to work properly. when i run its code to train the model it gives me it's error:

!keras_retinanet/bin/train.py \

--freeze-backbone
--random-transform
--weights {PRETRAINED_MODEL}
--batch-size 8
--steps 500
--epochs 10
csv annotations.csv classes.csv TypeError: type object got multiple values for keyword argument 'training'

the model does not train and i couldn't figure out the exact problem

1 Answers

Downgrading to keras==2.3.1 fixed the issue for me.

Related