Hi @CasiaFan I am getting the above error when I try t run the code with following error description.
[[Node: gradients/box_loss/Mul_3_grad/BroadcastGradientArgs = BroadcastGradientArgs[T=DT_INT32, _device="/job:localhost/replica:0/task:0/device:GPU:0"](box_loss/assert_broadcastable/values/shape, box_loss/assert_broadcastable/weights/shape)]]
[[Node: gradients/Retina_FCN_Head_Box/Repeat/convolution_4/BiasAdd_grad/BiasAddGrad/_1905 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device_incarnation=1, tensor_name="edge3196...iasAddGrad", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"]()]]
Hi @CasiaFan I am getting the above error when I try t run the code with following error description. [[Node: gradients/box_loss/Mul_3_grad/BroadcastGradientArgs = BroadcastGradientArgs[T=DT_INT32, _device="/job:localhost/replica:0/task:0/device:GPU:0"](box_loss/assert_broadcastable/values/shape, box_loss/assert_broadcastable/weights/shape)]] [[Node: gradients/Retina_FCN_Head_Box/Repeat/convolution_4/BiasAdd_grad/BiasAddGrad/_1905 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device_incarnation=1, tensor_name="edge3196...iasAddGrad", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"]()]]
Thank you.