Quickstart: Use tensors to compute train accuracy (#1662)

Co-authored-by: Dave Farago <[email protected]>
2023-08-24 17:09:12 -04:00
parent d37d092c14
commit 29adae84eb
1 changed files with 2 additions and 2 deletions
@@ -205,11 +205,11 @@ for step in range(1000):
  opt.step()

  # calculate accuracy
-  pred = out.argmax(axis=-1).numpy()
+  pred = out.argmax(axis=-1)
  acc = (pred == labels).mean()

  if step % 100 == 0:
-    print(f"Step {step+1} | Loss: {loss.numpy()} | Accuracy: {acc}")
+    print(f"Step {step+1} | Loss: {loss.numpy()} | Accuracy: {acc.numpy()}")
 ```

 ## Evaluation