# Assertion \`t \>= 0 && t \< n\_classes\` failed error

**URL:** <https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794>\
**Category:** vision\
**Created:** [October 9, 2021, 7:46am UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794 "2021-10-09T07:46:00Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![James\_Lee](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/james_lee/32/29487_2.png) [@James\_Lee](https://discuss.pytorch.org/u/James_Lee)\
**Post date:** [October 9, 2021, 7:46am UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794/1 "2021-10-09T07:46:00Z")

</div>

When I ran CUDA\_LAUNCH\_BLOCKING=1 python train.py I got the following error,

/opt/conda/conda-bld/pytorch\_1595629395347/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:106: cunn\_SpatialClassNLLCriterion\_updateOutput\_kernel: block: [0,0,0], thread: [342,0,0] Assertion `t >= 0 && t < n_classes` failed.  
/opt/conda/conda-bld/pytorch\_1595629395347/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:106: cunn\_SpatialClassNLLCriterion\_updateOutput\_kernel: block: [0,0,0], thread: [343,0,0] Assertion `t >= 0 && t < n_classes` failed.  
/opt/conda/conda-bld/pytorch\_1595629395347/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:106: cunn\_SpatialClassNLLCriterion\_updateOutput\_kernel: block: [0,0,0], thread: [344,0,0] Assertion `t >= 0 && t < n_classes` failed.  
/opt/conda/conda-bld/pytorch\_1595629395347/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:106: cunn\_SpatialClassNLLCriterion\_updateOutput\_kernel: block: [0,0,0], thread: [345,0,0] Assertion `t >= 0 && t < n_classes` failed.  
/opt/conda/conda-bld/pytorch\_1595629395347/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:106: cunn\_SpatialClassNLLCriterion\_updateOutput\_kernel: block: [0,0,0], thread: [346,0,0] Assertion `t >= 0 && t < n_classes` failed.  
/opt/conda/conda-bld/pytorch\_1595629395347/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:106: cunn\_SpatialClassNLLCriterion\_updateOutput\_kernel: block: [0,0,0], thread: [347,0,0] Assertion `t >= 0 && t < n_classes` failed.  
Traceback (most recent call last):  
File “train.py”, line 238, in  
main()  
File “train.py”, line 126, in main  
train(net, optimizer)  
File “train.py”, line 197, in train  
loss1 = criterion\_CE(out, torch.squeeze(labels).long())  
File “/home/public/software/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py”, line 722, in \_call\_impl  
result = self.forward(\*input, \*\*kwargs)  
File “/home/public/software/anaconda3/lib/python3.8/site-packages/torch/nn/modules/loss.py”, line 947, in forward  
return F.cross\_entropy(input, target, weight=self.weight,  
File “/home/public/software/anaconda3/lib/python3.8/site-packages/torch/nn/functional.py”, line 2422, in cross\_entropy  
return nll\_loss(log\_softmax(input, 1), target, weight, None, ignore\_index, None, reduction)  
File “/home/public/software/anaconda3/lib/python3.8/site-packages/torch/nn/functional.py”, line 2220, in nll\_loss  
ret = torch.\_C.\_nn.nll\_loss2d(input, target, weight, \_Reduction.get\_enum(reduction), ignore\_index)  
RuntimeError: cuda runtime error (710) : device-side assert triggered at /opt/conda/conda-bld/pytorch\_1595629395347/work/aten/src/THCUNN/generic/SpatialClassNLLCriterion.cu:134

It seemed that I should use nn.BCE rather than nn.CrossEntropyLoss here? I found only CrossEntropyLoss support ignore\_index for classification while nn.BCE didn’t.

Here’s my loss,

```auto
        criterion_CE = nn.CrossEntropyLoss(ignore_index=-1).cuda()
        loss = criterion_CE(out, torch.squeeze(labels).long())

```

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [October 9, 2021, 9:59am UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794/2 "2021-10-09T09:59:56Z")

</div>

The target tensor for `nn.CrossEntropyLoss` is expected to contain class indices in the range `[0, nb_classes-1]`, which seems to fail in your script.  
Check its values via `print(target.min(), target.max())` and make sure they are valid.

---

<div class="post-metadata">

**Author:** ![James\_Lee](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/james_lee/32/29487_2.png) [@James\_Lee](https://discuss.pytorch.org/u/James_Lee)\
**Post date:** [October 9, 2021, 10:59am UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794/3 "2021-10-09T10:59:33Z")

</div>

After `print(target.min(), target.max())` ,I got target.min() == 0 rather than -1 I expected. I set the label pixel value to -1 follow your advice here [Got RuntimeError: Boolean value of Tensor with more than one value is ambiguous during training - #2 by ptrblck](https://discuss.pytorch.org/t/got-runtimeerror-boolean-value-of-tensor-with-more-than-one-value-is-ambiguous-during-training/133762/2) , but it seemed this did not work as I expected? Could you please give me any clues? Many thanks.

---

<div class="post-metadata">

**Author:** ![James\_Lee](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/james_lee/32/29487_2.png) [@James\_Lee](https://discuss.pytorch.org/u/James_Lee)\
**Post date:** [October 9, 2021, 11:32am UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794/5 "2021-10-09T11:32:53Z")

</div>

Now I got the point. It’s this line of code lead to such strange behavior.

```
        labels1 = functional.interpolate(labels, size=24, mode='bilinear')

        print("### labels.long().min()", labels.long().min())
        print("### labels.min()", labels.min())

        print("### labels1.long().min()", labels1.long().min())
        print("### labels1.min()", labels1.min())

```

I got,

```auto
### labels.long().min() tensor(-1, device='cuda:2')
### labels.min() tensor(-1., device='cuda:2')
### labels1.long().min() tensor(0, device='cuda:2')
### labels1.min() tensor(0., device='cuda:2')

```

What’s wrong with functional.interpolate here? Thanks.

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [October 9, 2021, 8:13pm UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794/6 "2021-10-09T20:13:40Z")

</div>

You are interpolating values using the bilinear approach and rounding afterwards, which might change the values. I’m not familiar with your use case, but as previously described, the expected target values are in `[0, nb_classes-1]` unless you use `ignore_index` for a specific index value.  
Your current code crashes, because the target values are not in this range.

---

<div class="post-metadata">

**Author:** ![James\_Lee](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/james_lee/32/29487_2.png) [@James\_Lee](https://discuss.pytorch.org/u/James_Lee)\
**Post date:** [October 10, 2021, 2:55am UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794/7 "2021-10-10T02:55:24Z")

</div>

Thanks for your reply. Sorry to disturb you again.For the last question, I had set `ignore_index` to -1 .

criterion\_CE = nn.CrossEntropyLoss(ignore\_index=-1).cuda()

…  
labels[(0.3 \<= labels) & (labels \<= 0.7)] = -1  
…

loss = criterion\_CE(out, torch.squeeze(labels).long())

How could I fix this?

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [October 10, 2021, 10:29pm UTC](https://discuss.pytorch.org/t/assertion-t-0-t-n-classes-failed-error/133794/8 "2021-10-10T22:29:22Z")

</div>

I’m not sure what you are trying to fix.  
If you are manually setting some target indices to `-1`, which is invalid in the default setup, you have to use `ignore_index=-1`, since otherwise the expected error will be raised.