Open AI-liu opened 5 years ago
/opt/conda/conda-bld/pytorch_1535491974311/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:99: void cunn_SpatialClassNLLCriterion_updateOutput_kernel(T , T , T , long , T *, int, int, int, int, int, long) [with T = float, AccumT = float]: block: [4,0,0], thread: [607,0,0] Assertion t >= 0 && t < n_classes failed. THCudaCheck FAIL file=/opt/conda/conda-bld/pytorch_1535491974311/work/aten/src/THC/generic/THCTensorCopy.cpp line=70 error=59 : device-side assert triggered
t >= 0 && t < n_classes
Same error when trained on CityScapes, what is your environment?
maybe the labels of the images have questions. The labels should be labeltrainIds.png
/opt/conda/conda-bld/pytorch_1535491974311/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:99: void cunn_SpatialClassNLLCriterion_updateOutput_kernel(T , T , T , long , T *, int, int, int, int, int, long) [with T = float, AccumT = float]: block: [4,0,0], thread: [607,0,0] Assertion
t >= 0 && t < n_classes
failed. THCudaCheck FAIL file=/opt/conda/conda-bld/pytorch_1535491974311/work/aten/src/THC/generic/THCTensorCopy.cpp line=70 error=59 : device-side assert triggered