TFFRCNN
TFFRCNN copied to clipboard
Meet some problems when training on my own data. See below.
When the negative proposals is 0, the training ends up with the following complaints. Anyone met similar problems?
rpn: num_positive 128
rpn: num_negative 0
rpn: num_positive avg 128
rpn: num_negative avg 0
2018-03-26 15:54:27.734525: W tensorflow/core/framework/op_kernel.cc:1158] Unknown: exceptions.AssertionError: Invalid boxes found: [ 4.89326211e+04 -2.36359191e+00 -8.73799973e+01 1.20674377e+02] [ 439.95526123 0. 618.95306396 49.03554916]
[[Node: roi-data/PyFunc = PyFunc[Tin=[DT_FLOAT, DT_FLOAT, DT_INT32, DT_FLOAT, DT_INT32], Tout=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], token="pyfunc_2", _device="/job:localhost/replica:0/task:0/cpu:0"](rpn_rois/_97, _arg_gt_boxes_0_3, _arg_gt_ishard_0_4, _arg_dontcare_areas_0_2, roi-data/PyFunc/input_4)]]
2018-03-26 15:54:27.734525: W tensorflow/core/framework/op_kernel.cc:1158] Unknown: exceptions.AssertionError: Invalid boxes found: [ 4.89326211e+04 -2.36359191e+00 -8.73799973e+01 1.20674377e+02] [ 439.95526123 0. 618.95306396 49.03554916]
[[Node: roi-data/PyFunc = PyFunc[Tin=[DT_FLOAT, DT_FLOAT, DT_INT32, DT_FLOAT, DT_INT32], Tout=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], token="pyfunc_2", _device="/job:localhost/replica:0/task:0/cpu:0"](rpn_rois/_97, _arg_gt_boxes_0_3, _arg_gt_ishard_0_4, _arg_dontcare_areas_0_2, roi-data/PyFunc/input_4)]]
2018-03-26 15:54:27.734533: W tensorflow/core/framework/op_kernel.cc:1158] Unknown: exceptions.AssertionError: Invalid boxes found: [ 4.89326211e+04 -2.36359191e+00 -8.73799973e+01 1.20674377e+02] [ 439.95526123 0. 618.95306396 49.03554916]
[[Node: roi-data/PyFunc = PyFunc[Tin=[DT_FLOAT, DT_FLOAT, DT_INT32, DT_FLOAT, DT_INT32], Tout=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], token="pyfunc_2", _device="/job:localhost/replica:0/task:0/cpu:0"](rpn_rois/_97, _arg_gt_boxes_0_3, _arg_gt_ishard_0_4, _arg_dontcare_areas_0_2, roi-data/PyFunc/input_4)]]
do you solve the problem?I have the same question ,you can keep in touch with me [email protected]
I have the same question
Did you solve this problem? I have the same question.