Traceback (most recent call last):
│ File "samples/coco/coco.py", line 504, in <module>
│ augmentation=augmentation)
│ File "/home/alex/anaconda3/lib/python3.7/site-packages/mask_rcnn-2.1-py3.7.egg/mrcnn/model.py", line 2354, in train
│ File "/home/alex/anaconda3/lib/python3.7/site-packages/mask_rcnn-2.1-py3.7.egg/mrcnn/model.py", line 2199, in compile
│AttributeError: 'Model' object has no attribute 'metrics_tensors'
you might try to find the suitable combination experimentally. For me worked: keras 2.2.0, tf 1.6.0, python3.6. Another nice question is - how to make it parallel?
I am trying train the mask-rcnn using coco dataset:
But it gave me the follows error:
My keras version is 2.3.0.