1.
TypeError: 'zip' object is not subscriptable
λ python gen_model.py -s train -c 6 >train.prototxt
Traceback (most recent call last):
File "gen_model.py", line 891, in <module>
gen.generate(FLAGS.stage, not FLAGS.classifier, FLAGS.size, FLAGS.class_num, not FLAGS.batchnorm, FLAGS.eps, FLAGS.relu6)
File "gen_model.py", line 808, in generate
self.mbox("conv_13/expand", 576, 3)
File "gen_model.py", line 711, in mbox
min_size, max_size = self.anchors[0]
TypeError: 'zip' object is not subscriptable
解决方法: gen_model.py的第834行, 把
return zip(scales[:-1], scales[1:])
改为
return list(zip(scales[:-1], scales[1:]))
2.
Traceback (most recent call last):
File "dump_tensorflow_weights.py", line 95, in <module>
tmp = caffe_weights.reshape(boxes, -1).copy()
TypeError: 'float' object cannot be interpreted as an integer
解决方法:
将dump_tensorflow_weights.py的第74,74,114,116行的"/"改为"//"