[doc] Reset curves after changes to plotting strategy

doc/references.rst

@@ -104,3 +104,7 @@
 .. [SANDLER-2018] *M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, L.-C.h Chen*,
    **MobileNetV2: Inverted Residuals and Linear Bottlenecks**, 2018.
    https://arxiv.org/abs/1801.04381
+
+.. [DAVIS-2006] *J. Davis and M. Goadrich*, **The relationship between
+   Precision-Recall and ROC curves**. 23rd international conference on Machine
+   learning (ICML’06), 2006. https://doi.org/10.1145/1143844.1143874

doc/results/baselines/index.rst

@@ -82,7 +82,25 @@ Next, you will find the PR plots showing confidence intervals, for the various
 models explored, on a per dataset arrangement.  All curves correspond to test
 set performances.  Single performance figures (F1-micro scores) correspond to
 its average value across all test set images, for a fixed threshold set to
-``0.5``.
+``0.5``, and using 1000 points for curve calculation.
+
+.. tip:: **Curve Intepretation**
+
+   PR curves behave differently than traditional ROC curves (using Specificity
+   versus Sensitivity) with respect to the overall shape.  You may have a look
+   at [DAVIS-2006]_ for details on the relationship between PR and ROC curves.
+   For example, PR curves are not guaranteed to be monotonically increasing or
+   decreasing with the scanned thresholds (e.g. see M2U-Net on STARE dataset).
+
+   Each evaluated threshold in a combination of trained models and datasets is
+   represented by a point in each curve.  Points are linearly interpolated to
+   created a line.  For each evaluated threshold and every trained model and
+   dataset, we assume that the standard deviation on both precision and recall
+   estimation represent good proxies for the uncertainty around that point.  We
+   therefore plot a transparent ellipse centered around each evaluated point in
+   which the width corresponds to twice the recall standard deviation and the
+   height, twice the precision standard deviation.
+
 
 .. list-table::
 
@@ -133,4 +151,5 @@ Remarks
   models show consistently less variability than the second annotator.
   Unfortunately, this cannot be conclusive.
 
+
 .. include:: ../../links.rst

doc/results/xtest/index.rst

@@ -94,7 +94,7 @@ cross-tests explored, on a per cross-tested model arrangement.  All curves
 correspond to test set performances.  Single performance figures (F1-micro
 scores) correspond to its average value across all test set images, for a fixed
 threshold set *a priori* on the training set of dataset used for creating the
-model.
+model, and using 100 points for curve calculation.
 
 .. list-table::