<p>I'm posting this question for those who made the same error I did. I got this error when trying to compute my gradients:</p> <pre class="prettyprint"><code>criterion = torch.nn.CrossEntropyLoss() loss = criterion(y_hat, y_truth) loss.backwards() </code></pre>

<p>It's <code>loss.backward()</code>, not <code>loss.backwards()</code>.</p>

AttributeError: 'Tensor' has no attribute: 'backwards'

I'm posting this question for those who made the same error I did. I got this error when trying to compute my gradients:

criterion = torch.nn.CrossEntropyLoss()
loss = criterion(y_hat, y_truth)
loss.backwards()

574

asked Oct 15 '18 12:10

It's loss.backward(), not loss.backwards().

answered Oct 19 '22 23:10

If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!