Skip to content

Commit 4ccdf99

Browse files
author
Philip Gautier
committed
Fix typos in linear learner multiclass notebook
1 parent 9eab44e commit 4ccdf99

File tree

1 file changed

+2
-2
lines changed

1 file changed

+2
-2
lines changed

scientific_details_of_algorithms/linear_learner_multiclass_classification/linear_learner_multiclass_classification.ipynb

+2-2
Original file line numberDiff line numberDiff line change
@@ -246,7 +246,7 @@
246246
"cell_type": "markdown",
247247
"metadata": {},
248248
"source": [
249-
"We'll add a convenience function for parsing predictions and evaluating model metrics. It will feed test features to the endpoint and receive predicted test labels. To evaluate the models we create, we'll capture predicted test labels and compare them to actuals using some common multiclass classification metrics. As mentioned earlier, we're extracting the `predicted_label` from each response payload. That's the class with the highest predicted probability. We'll get one class label per example. To get a vector of six probabilities for each example (the predicted probability for each class) , we would extract the `score` from the response payload. Details of linear learner's response format are in the [documentation](https://docs.aws.amazon.com/sagemaker/latest/dg/LL-in-formats.html)."
249+
"We'll add a convenience function for parsing predictions and evaluating model metrics. It will feed test features to the endpoint and receive predicted test labels. To evaluate the models we create, we'll capture predicted test labels and compare them to actuals using some common multiclass classification metrics. As mentioned earlier, we're extracting the `predicted_label` from each response payload. That's the class with the highest predicted probability. We'll get one class label per example. To get a vector of seven probabilities for each example (the predicted probability for each class) , we would extract the `score` from the response payload. Details of linear learner's response format are in the [documentation](https://docs.aws.amazon.com/sagemaker/latest/dg/LL-in-formats.html)."
250250
]
251251
},
252252
{
@@ -407,7 +407,7 @@
407407
"metadata": {},
408408
"outputs": [],
409409
"source": [
410-
"# evaluate metrics of the model trained with default hyperparameters\n",
410+
"# evaluate metrics of the model trained with balanced class weights\n",
411411
"evaluate_metrics(balanced_multiclass_predictor, test_features, test_labels)"
412412
]
413413
},

0 commit comments

Comments
 (0)