doc: updated gh-pages

deel-ai · Oct 3, 2023 · 963cab5 · 963cab5
1 parent 79ea0b9
commit 963cab5
Show file tree

Hide file tree

Showing 84 changed files with 6,861 additions and 11 deletions.
diff --git a/docs/.buildinfo b/docs/.buildinfo
@@ -1,4 +1,4 @@
 # Sphinx build info version 1
 # This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done.
-config: 1d10e2497f8b5c2e668e5b79c4cf04e7
+config: 5f0987fab81c15202f538c36011e1bf5
 tags: 645f666f9bcd5a90fca523b33c5a78b7
diff --git a/docs/_sources/anomaly_detection.rst.txt b/docs/_sources/anomaly_detection.rst.txt
@@ -0,0 +1,15 @@
+.. _anomaly_detection:
+
+🚩 Anomaly detection
+====================
+
+
+Currently implemented conformal anomaly detectors are listed in this page.
+
+Each of these wrappers calibrate the decision threshold for anomaly detectors 
+that are passed as argument in the object constructor. Such models **need** to
+implement the :func:`fit` and :func:`predict` methods.
+:doc:`Prediction module <prediction>` from the :doc:`API <api>` ensures the
+compliance of models from various ML/DL libraries (such as Keras and scikit-learn) to **puncc**.
+
+.. autoclass:: deel.puncc.anomaly_detection.SplitCAD
diff --git a/docs/_sources/getting_started.rst.txt b/docs/_sources/getting_started.rst.txt
@@ -16,13 +16,13 @@ We want to evaluate the uncertainty associated with the prediction using **induc
 💾 Diabetes Dataset
 *******************
 
-The dataset contains information about 442 diabetes patients. The goal is 
-predict from physiological variables a quantitative measure of disease 
+The dataset contains information about 442 diabetes patients. The goal is
+predict from physiological variables a quantitative measure of disease
 progression in one year.
 
 * There are ten standardized features corresponding to the age, sex, body mass
-  index, average blood pressure, and six blood serum measurements. 
-* The target is the measure of diabetes progression during one year for each 
+  index, average blood pressure, and six blood serum measurements.
+* The target is the measure of diabetes progression during one year for each
   patient.
 
 for more information, check the official
@@ -44,12 +44,12 @@ for more information, check the official
    Features shape: (442, 10)
    Target's shape: (442,)
 
-From all the features, we want our model to capture only the link between 
-body mass index and the evolution of the disease. 
+From all the features, we want our model to capture only the link between
+body mass index and the evolution of the disease.
 
 .. code-block:: python
 
-   import numpy as np 
+   import numpy as np
 
    # Use only BMI feature
    diabetes_X = diabetes_X[:, 2, np.newaxis]
@@ -366,4 +366,4 @@ and the average empirical size of the prediction sets on the test examples:
 .. code-block:: HTML
 
    Empirical coverage : 0.90
-   Average set size : 1.03
+   Average set size : 1.03
diff --git a/docs/_sources/index.rst.txt b/docs/_sources/index.rst.txt
@@ -16,6 +16,7 @@ true outputs within a user-defined error :math:`\alpha`.
    getting_started
    regression
    classification
+   anomaly_detection
    api
    metrics
    plotting

diff --git a/docs/_sources/theory_overview.rst.txt b/docs/_sources/theory_overview.rst.txt
@@ -173,7 +173,7 @@ If :math:`K = n`, we obtain the *Jackknife+*, **leave-one-out** version of the a
 
 .. Let :math:`(X_{new}, Y_{new})` be a test point, where :math:`Y_{new}` is not observable at inference time.
 
-The lower and upper bounds of the prediction interval are given by: 
+The lower and upper bounds of the prediction interval are given by:
 
     1. Compute :math:`\bar{R}_{L} = \{ \widehat{f}_{-S_{k(i)}}(X_{new}) - R_i^{CV} \}_{i=1}^{n}`
     2. :math:`\widehat{L}_{\alpha}(X_{new}) = \lfloor \alpha (n+1) \rfloor`-th smallest value in :math:`\bar{R}_{L}` (lower bound)
@@ -237,6 +237,11 @@ Regularized Adaptive Prediction Sets (RAPS)
 
 TBC
 
+Conformal Anomaly Detection
+---------------------------
+
+TBC
+
 References
 ----------
 

diff --git a/docs/anomaly_detection.html b/docs/anomaly_detection.html
diff --git a/docs/build/doctrees/anomaly_detection.doctree b/docs/build/doctrees/anomaly_detection.doctree
diff --git a/docs/build/doctrees/api.doctree b/docs/build/doctrees/api.doctree
diff --git a/docs/build/doctrees/calibration.doctree b/docs/build/doctrees/calibration.doctree
diff --git a/docs/build/doctrees/classification.doctree b/docs/build/doctrees/classification.doctree
diff --git a/docs/build/doctrees/conformalization.doctree b/docs/build/doctrees/conformalization.doctree
diff --git a/docs/build/doctrees/environment.pickle b/docs/build/doctrees/environment.pickle
diff --git a/docs/build/doctrees/getting_started.doctree b/docs/build/doctrees/getting_started.doctree
diff --git a/docs/build/doctrees/index.doctree b/docs/build/doctrees/index.doctree
diff --git a/docs/build/doctrees/metrics.doctree b/docs/build/doctrees/metrics.doctree
diff --git a/docs/build/doctrees/nonconformity_scores.doctree b/docs/build/doctrees/nonconformity_scores.doctree
diff --git a/docs/build/doctrees/plotting.doctree b/docs/build/doctrees/plotting.doctree
diff --git a/docs/build/doctrees/prediction.doctree b/docs/build/doctrees/prediction.doctree
diff --git a/docs/build/doctrees/prediction_sets.doctree b/docs/build/doctrees/prediction_sets.doctree
diff --git a/docs/build/doctrees/regression.doctree b/docs/build/doctrees/regression.doctree
diff --git a/docs/build/doctrees/splitting.doctree b/docs/build/doctrees/splitting.doctree
diff --git a/docs/build/doctrees/theory_overview.doctree b/docs/build/doctrees/theory_overview.doctree
diff --git a/docs/build/doctrees/utils.doctree b/docs/build/doctrees/utils.doctree
diff --git a/docs/build/html/.buildinfo b/docs/build/html/.buildinfo
@@ -0,0 +1,4 @@
+# Sphinx build info version 1
+# This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done.
+config: 5f0987fab81c15202f538c36011e1bf5
+tags: 645f666f9bcd5a90fca523b33c5a78b7
diff --git a/docs/build/html/.nojekyll b/docs/build/html/.nojekyll
diff --git a/docs/build/html/_images/k-fold-scheme.png b/docs/build/html/_images/k-fold-scheme.png
diff --git a/docs/build/html/_images/results_quickstart_aps_mnist.png b/docs/build/html/_images/results_quickstart_aps_mnist.png
diff --git a/docs/build/html/_images/results_quickstart_split_cp_pi.png b/docs/build/html/_images/results_quickstart_split_cp_pi.png
diff --git a/docs/build/html/_sources/anomaly_detection.rst.txt b/docs/build/html/_sources/anomaly_detection.rst.txt
@@ -0,0 +1,15 @@
+.. _anomaly_detection:
+
+🚩 Anomaly detection
+====================
+
+
+Currently implemented conformal anomaly detectors are listed in this page.
+
+Each of these wrappers calibrate the decision threshold for anomaly detectors 
+that are passed as argument in the object constructor. Such models **need** to
+implement the :func:`fit` and :func:`predict` methods.
+:doc:`Prediction module <prediction>` from the :doc:`API <api>` ensures the
+compliance of models from various ML/DL libraries (such as Keras and scikit-learn) to **puncc**.
+
+.. autoclass:: deel.puncc.anomaly_detection.SplitCAD