Single Dataset Performance#

This notebooks provides an overview for using and understanding single dataset performance check.

Structure:

What Is the Purpose of the Check?#

This check returns the results from a dict of metrics, in the format metric name: scorer, calculated for the given model dataset. The scorer should be either a sklearn scorer or a custom metric (see Metrics Guide for further details). Use this check to evaluate the performance on a single vision dataset such as a test set.

Generate Dataset#

Note

In this example, we use the pytorch version of the mnist dataset and model. In order to run this example using tensorflow, please change the import statements to:

from deepchecks.vision.datasets.classification import mnist_tensorflow as mnist
from deepchecks.vision.checks import SingleDatasetPerformance
from deepchecks.vision.datasets.classification import mnist_torch as mnist
train_ds = mnist.load_dataset(train=True, object_type='VisionData')

Run the check#

The check will use the default classification metrics - precision and recall.

check = SingleDatasetPerformance()
result = check.run(train_ds)
result.show()
Processing Batches:
|     | 0/1 [Time: 00:00]
Processing Batches:
|█████| 1/1 [Time: 00:01]
Processing Batches:
|█████| 1/1 [Time: 00:01]

Computing Check:
|     | 0/1 [Time: 00:00]

Computing Check:
|█████| 1/1 [Time: 00:00]
Single Dataset Performance


To display the results in an IDE like PyCharm, you can use the following code:

#  result.show_in_window()

The result will be displayed in a new window.

Now we will run a check with a metric different from the defaults- F-1. You can read more about setting metrics in the Metrics Guide.

check = SingleDatasetPerformance(scorers={'f1': 'f1_per_class'})
result = check.run(train_ds)
result
Processing Batches:
|     | 0/1 [Time: 00:00]
Processing Batches:
|█████| 1/1 [Time: 00:01]
Processing Batches:
|█████| 1/1 [Time: 00:01]

Computing Check:
|     | 0/1 [Time: 00:00]

Computing Check:
|█████| 1/1 [Time: 00:00]
Single Dataset Performance


Define a Condition#

We can define a condition to validate that our model performance score is above or below a certain threshold. The condition is defined as a function that takes the results of the check as input and returns a ConditionResult object.

check = SingleDatasetPerformance()
check.add_condition_greater_than(0.5)
result = check.run(train_ds)
result.show(show_additional_outputs=False)
Processing Batches:
|     | 0/1 [Time: 00:00]
Processing Batches:
|█████| 1/1 [Time: 00:01]
Processing Batches:
|█████| 1/1 [Time: 00:01]

Computing Check:
|     | 0/1 [Time: 00:00]

Computing Check:
|█████| 1/1 [Time: 00:00]
Single Dataset Performance


We can also define a condition on a specific metric (or a subset of the metrics) that was passed to the check and a specific class, instead of testing all the metrics and all the classes which is the default mode.

check = SingleDatasetPerformance()
check.add_condition_greater_than(0.8, metrics=['Precision'], class_mode='3')
result = check.run(train_ds)
result.show(show_additional_outputs=False)
Processing Batches:
|     | 0/1 [Time: 00:00]
Processing Batches:
|█████| 1/1 [Time: 00:01]
Processing Batches:
|█████| 1/1 [Time: 00:01]

Computing Check:
|     | 0/1 [Time: 00:00]

Computing Check:
|█████| 1/1 [Time: 00:00]
Single Dataset Performance


Total running time of the script: (0 minutes 5.571 seconds)

Gallery generated by Sphinx-Gallery