ThomasDelteil commented on a change in pull request #9583: use nd for accuracy
calculation
URL: https://github.com/apache/incubator-mxnet/pull/9583#discussion_r175591965
##########
File path: python/mxnet/metric.py
##########
@@ -380,23 +380,27 @@ def update(self, labels, preds):
Parameters
----------
labels : list of `NDArray`
- The labels of the data.
+ The labels of the data with class indices as values, one per
sample.
preds : list of `NDArray`
- Predicted values.
+ Prediction values for samples. Each prediction value can either be
the class index,
+ or a vector of likelihoods for all classes.
"""
check_label_shapes(labels, preds)
for label, pred_label in zip(labels, preds):
if pred_label.shape != label.shape:
pred_label = ndarray.argmax(pred_label, axis=self.axis)
- pred_label = pred_label.asnumpy().astype('int32')
- label = label.asnumpy().astype('int32')
+ pred_label = pred_label.astype('int32')
+ label = label.astype('int32')
check_label_shapes(label, pred_label)
- self.sum_metric += (pred_label.flat == label.flat).sum()
- self.num_inst += len(pred_label.flat)
+ if pred_label.context != label.context:
+ pred_label = pred_label.as_in_context(label.context)
+
+ self.sum_metric += (pred_label.flatten() ==
label.flatten()).sum().asscalar()
Review comment:
Maybe my understanding is flawed but in order to be able to bring back this
value on the CPU to self.sum_metric, we will need to do a wait_on_read on the
pred_label ?
If we have a loop that has:
```
for batch in data:
...
metric.update()
```
we will wait for the label_pred to be computed before we load the next batch
on the GPU.
Whilst if we had `self.sum_metric = nd.zeros(1, ctx=ctx)`
`self.sum_metric += (pred_label.flatten() == label.flatten()).sum()`
then we wouldn't be blocking on this operation before loading the next batch
of data and it would just enqueue more operations on the back-end.
I have run several experiments where having the loss computed .asscalar() on
every loop is slowing the training by 2-25%.
see:
https://github.com/ilkarman/DeepLearningFrameworks/issues/55
https://github.com/zackchase/mxnet-the-straight-dope/issues/455
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
