# Cross Entropy Loss for a Sequence (Time Series) of Output? **URL:** \ **Category:** Uncategorized\ **Created:** [June 26, 2017, 1:01am UTC](https://discuss.pytorch.org/t/cross-entropy-loss-for-a-sequence-time-series-of-output/4309 "2017-06-26T01:01:43Z")\ **Posts on this page:** 4\ **Page:** 1
**Author:** ![dlmacedo](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/dlmacedo/32/1387_2.png) [@dlmacedo](https://discuss.pytorch.org/u/dlmacedo)\ **Post date:** [June 26, 2017, 1:01am UTC](https://discuss.pytorch.org/t/cross-entropy-loss-for-a-sequence-time-series-of-output/4309/1 "2017-06-26T01:01:43Z")
The output of my network is a tensor of size torch.Size([time\_steps, 20, 29]). 20 is the batch size, and 29 is the number of classes. time\_steps is variable and depends on the input. My targets has the form torch.Size([time\_steps, 20]). How can I calculate the loss using nn.CrossEntropyLoss function? It should be noticed that the loss should be the sum of the loss of time\_steps cross entropy loss operations. ---
**Author:** ![chenyuntc](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/chenyuntc/32/299_2.png) [@chenyuntc](https://discuss.pytorch.org/u/chenyuntc)\ **Post date:** [June 26, 2017, 6:22am UTC](https://discuss.pytorch.org/t/cross-entropy-loss-for-a-sequence-time-series-of-output/4309/2 "2017-06-26T06:22:26Z")
```python output = output.view(-1,29) target = target.view(-1) criterion = nn.CrossEntropyLoss(29) loss = criterion(output,target) ``` ---
**Author:** ![dlmacedo](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/dlmacedo/32/1387_2.png) [@dlmacedo](https://discuss.pytorch.org/u/dlmacedo)\ **Post date:** [June 26, 2017, 11:55pm UTC](https://discuss.pytorch.org/t/cross-entropy-loss-for-a-sequence-time-series-of-output/4309/4 "2017-06-26T23:55:12Z")
Why do I need pass 29 to CrossEntropyLoss? Could I just use nn.CrossEntropyLoss()? ---
**Author:** ![DuaneNielsen](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/duanenielsen/32/4158_2.png) [@DuaneNielsen](https://discuss.pytorch.org/u/DuaneNielsen)\ **Post date:** [May 13, 2018, 8:59pm UTC](https://discuss.pytorch.org/t/cross-entropy-loss-for-a-sequence-time-series-of-output/4309/5 "2018-05-13T20:59:23Z")
For pytorch 3,this worked fine for me output = output.view(-1,29) target = target.view(-1) criterion = nn.CrossEntropyLoss() loss = criterion(output,target)