How long does it take to train? Has anyone trained successfully?

[ghost]

[ghost]

I found the calculation of edges costs most time in training.

[BestSongEver]

I tried to run the training on my own dataset, but got an issue: ZeroDivisionError: float division by zero

UnknownError (see above for traceback): exceptions.ZeroDivisionError: float division by zero [[Node: edges/PyFunc = PyFunc[Tin=[DT_FLOAT, DT_FLOAT], Tout=[DT_FLOAT], token="pyfunc_4", _device="/job:localhost/replica:0/task:0/cpu:0"](roi-data/rois/_141, _arg_Placeholder_1_0_1)]] [[Node: edges/PyFunc/_143 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_1461_edges/PyFunc", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/gpu:0"]]

Still confusing... Could u please help?

[ghost]

I tried to run the training on my own dataset, but got an issue: ZeroDivisionError: float division by zero

UnknownError (see above for traceback): exceptions.ZeroDivisionError: float division by zero [[Node: edges/PyFunc = PyFunc[Tin=[DT_FLOAT, DT_FLOAT], Tout=[DT_FLOAT], token="pyfunc_4", _device="/job:localhost/replica:0/task:0/cpu:0"](roi-data/rois/_141, _arg_Placeholder_1_0_1)]] [[Node: edges/PyFunc/_143 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_1461_edges/PyFunc", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/gpu:0"]]

Still confusing... Could u please help? emmm, I re-implemented it with Pytorch and training it on VOC2007. There is not error when training except the speed is so slow. Maybe you could locate your error and add a small constant.

[Elasine]

No, I met the problem with incompatible shape.So sad.

[zhl95]

No, I met the problem with incompatible shape.So sad.

Hi Elasine, I have the same problem as you. Did you have a clue to solve it?