Skip to content

Commit f49202c

Browse files
committed
updating MCM docs
1 parent 247dfe7 commit f49202c

File tree

7 files changed

+59
-8440
lines changed

7 files changed

+59
-8440
lines changed

docs/_static/tutorials/performance_estimation/multiclass/tutorial-confusion-matrix-estimation-multiclass-analysis-with-ref.svg

+1-1
Loading

docs/example_notebooks/Tutorial - Estimating Confusion Matrix - Multiclass Classification.ipynb

+39-8,418
Large diffs are not rendered by default.

docs/tutorials/performance_calculation/multiclass_performance_calculation/confusion_matrix_calculation.rst

+2-2
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
11
.. _multiclass-confusion-matrix-calculation:
22

3-
========================================================================================
3+
===================================================================
44
Calculating Confusion Matrix Elements for Multiclass Classification
5-
========================================================================================
5+
===================================================================
66

77
This tutorial explains how to use NannyML to calculate the :term:`confusion matrix<Confusion Matrix>` for multiclass classification
88
models.

docs/tutorials/performance_calculation/multiclass_performance_calculation/standard_metric_calculation.rst

+3-3
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
11
.. _multiclass-standard-metric-calculation:
22

3-
================================================================
4-
Monitoring Realized Performance for Multiclass Classification
5-
================================================================
3+
======================================================================
4+
Calculating Standard Performance Metrics for Multiclass Classification
5+
======================================================================
66

77
.. note::
88
The following example uses :term:`timestamps<Timestamp>`.

docs/tutorials/performance_estimation/multiclass_performance_estimation.rst

+2-2
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
11
.. _multiclass-performance-estimation:
22

3-
================================================
3+
====================================================
44
Estimating Performance for Multiclass Classification
5-
================================================
5+
====================================================
66

77
We currently support the following **standard** metrics for multiclass classification performance estimation:
88

docs/tutorials/performance_estimation/multiclass_performance_estimation/confusion_matrix_estimation.rst

+6-7
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,9 @@
11
.. _multiclass-confusion-matrix-estimation:
22

3-
========================================================================================
3+
==================================================================
44
Estimating Confusion Matrix Elements for Multiclass Classification
5-
========================================================================================
5+
==================================================================
6+
67
This tutorial explains how to use NannyML to estimate the :term:`confusion matrix<Confusion Matrix>` for multiclass classification
78
models in the absence of target data. To find out how CBPE estimates performance, read the :ref:`explanation of Confidence-based
89
Performance Estimation<performance-estimation-deep-dive>`.
@@ -162,8 +163,6 @@ What's next
162163
-----------
163164

164165
The :ref:`Data Drift<data-drift>` functionality can help us to understand whether data drift is causing the performance problem.
165-
166-
You can learn more about the Confidence Based Performance Estimation and its limitations in the
167-
:ref:`How it Works page<performance-estimation-deep-dive>`.
168-
169-
166+
When the target values become available we can use
167+
:ref:`realized performance calculation<multiclass-confusion-matrix-calculation>` to
168+
:ref:`compare realized and estimated confusion matrix results<compare_estimated_and_realized_performance>`.

docs/tutorials/performance_estimation/multiclass_performance_estimation/standard_metric_estimation.rst

+6-7
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
11
.. _multiclass_standard-metric-estimation:
22

3-
====================================================
4-
Estimating Performance for Multiclass Classification
5-
====================================================
3+
=====================================================================
4+
Estimating Standard Performance Metrics for Multiclass Classification
5+
=====================================================================
66

77
This tutorial explains how to use NannyML to estimate the performance of binary classification
88
models in the absence of target data. To find out how :class:`~nannyml.performance_estimation.confidence_based.cbpe.CBPE` estimates performance, read the :ref:`explanation of Confidence-based
@@ -140,7 +140,6 @@ What's next
140140
-----------
141141

142142
The :ref:`Data Drift<data-drift>` functionality can help us to understand whether data drift is causing the performance problem.
143-
When the target results become available they can be :ref:`compared with the estimated results<compare_estimated_and_realized_performance>`.
144-
145-
You can learn more about the Confidence Based Performance Estimation and its limitations in the
146-
:ref:`How it Works page<performance-estimation-deep-dive>`.
143+
When the target values become available we can use
144+
:ref:`realized performance calculation<multiclass-standard-metric-calculation>` to
145+
:ref:`compare realized and estimated confusion matrix results<compare_estimated_and_realized_performance>`.

0 commit comments

Comments
 (0)