search

Tramac / awesome-semantic-segmentation-pytorch

Semantic Segmentation on PyTorch (include FCN, PSPNet, Deeplabv3, Deeplabv3+, DANet, DenseASPP, BiSeNet, EncNet, DUNet, ICNet, ENet, OCNet, CCNet, PSANet, CGNet, ESPNet, LEDNet, DFANet)
Apache License 2.0
2.82k stars 581 forks source link

111 #83

Open anqin5211314 opened 4 years ago

anqin5211314 commented 4 years ago

/pytorch/aten/src/THCUNN/SpatialClassNLLCriterion.cu:103: void cunn_SpatialClassNLLCriterion_updateOutput_kernel(T , T , T , long , T , int, int, int, int, int, long) [with T = float, AccumT = float]: block: [0,0,0], thread: [987,0,0] Assertion t >= 0 && t < n_classes failed. /pytorch/aten/src/THCUNN/SpatialClassNLLCriterion.cu:103: void cunn_SpatialClassNLLCriterion_updateOutput_kernel(T , T , T , long , T , int, int, int, int, int, long) [with T = float, AccumT = float]: block: [0,0,0], thread: [988,0,0] Assertion t >= 0 && t < n_classes failed. /pytorch/aten/src/THCUNN/SpatialClassNLLCriterion.cu:103: void cunn_SpatialClassNLLCriterion_updateOutput_kernel(T , T , T , long , T , int, int, int, int, int, long) [with T = float, AccumT = float]: block: [0,0,0], thread: [989,0,0] Assertion t >= 0 && t < n_classes failed. /pytorch/aten/src/THCUNN/SpatialClassNLLCriterion.cu:103: void cunn_SpatialClassNLLCriterion_updateOutput_kernel(T , T , T , long , T , int, int, int, int, int, long) [with T = float, AccumT = float]: block: [0,0,0], thread: [990,0,0] Assertion t >= 0 && t < n_classes failed. /pytorch/aten/src/THCUNN/SpatialClassNLLCriterion.cu:103: void cunn_SpatialClassNLLCriterion_updateOutput_kernel(T , T , T , long , T *, int, int, int, int, int, long) [with T = float, AccumT = float]: block: [0,0,0], thread: [991,0,0] Assertion t >= 0 && t < n_classes failed. Traceback (most recent call last): File "train.py", line 331, in trainer.train() File "train.py", line 235, in train losses.backward() File "/home/anqin/anaconda3/envs/ICnet/lib/python3.6/site-packages/torch/tensor.py", line 107, in backward torch.autograd.backward(self, gradient, retain_graph, create_graph) File "/home/anqin/anaconda3/envs/ICnet/lib/python3.6/site-packages/torch/autograd/init.py", line 93, in backward allow_unreachable=True) # allow_unreachable flag RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED

sainatarajan commented 4 years ago

This error tells that some of the class values in the pixels of your masks are greater than n_classes. Try this:

img= cv2.imread(path_to_mask)
print(numpy.unique(img, return_counts= True))

This will show the unique pixel values and their counts. Check if you have any pixel values more than n_classes.