project_5_vehicle_detection -- Got TypeError: 'module' object is not callable when run main_ssd.py
binlidaily opened this issue · 1 comments
binlidaily commented
When I run main_ssd.py I got this:
Traceback (most recent call last):
File "/home/binli/self-driving-car/project_5_vehicle_detection/main_ssd.py", line 10, in <module>
ssd_model, bbox_helper, color_palette = get_SSD_model()
File "/home/binli/self-driving-car/project_5_vehicle_detection/SSD.py", line 805, in get_SSD_model
model_ssd = SSD300(input_shape=(300, 300, 3), num_classes=NUM_CLASSES, pretrained=True)
File "/home/binli/self-driving-car/project_5_vehicle_detection/SSD.py", line 679, in SSD300
mode='concat', concat_axis=1, name='mbox_loc')
TypeError: 'module' object is not callable
I tried a lot but I cannot fix this? Any suggestions about this? @ndrplz thanks a lot
ndrplz commented
Not at the pc right now, but from the stack trace it looks like an issue of the merge function.
Take a look that the one in the code is still a valid way for merging activations. The keras version is now legacy