docs: improve documention about classifications

docs/en/functionalities/classification.md

@@ -1,13 +1,5 @@
 # Classification panel
 
-::: warning
-
-The english version of the documentation is currently under construction.
-
-Some parts are not yet translated and some translations may be incomplete or inaccurate.
-
-:::
-
 Several methods are proposed for transforming a continuous series of values into a discrete series, i.e. into a finite number of classes.
 
 When creating a choropleth representation, the number of classes and their limit values must be justified statistically and/or thematically.
@@ -16,6 +8,12 @@ The methods proposed by the tool can be used as they stand, or as reading and an
 
 ## Overview of the classification panel
 
+<ZoomImg
+    src="../classification.png"
+    alt="Classification panel"
+    caption="Classification panel"
+/>
+
 Several elements are present in this window:
 
 - A summary of the series of values to be classified (number of non-zero values, mean, median, minimum, maximum, etc.),
@@ -36,6 +34,7 @@ This method, sometimes also called "equal amplitudes", allows the creation of cl
 
 This method (popularized by the PhilCarto tool), allows classification according to the quartile method while isolating extreme values:
 it thus produces 6 classes.
+The 6 classes are defined with the following limits: minimum, percentile 5 (0.05%), 1st quartile (25%), median (50%), 3rd quartile (75%), percentile 95 (95%), maximum.
 
 ## Natural thresholds (CKMeans algorithm)
 
@@ -60,14 +59,35 @@ This method of classification does not allow you to directly choose a number of
 which corresponds to the size of a class as well as the role of the mean (used as a class boundary or
 as a class center).
 
+### Geometric progression
+
+This classification method allows you to create classes whose limits are defined by a geometric progression: each class is defined by a multiple of the previous one.
+
 ### Head/tail breaks
 
+This classification method can be used to create classes for series that are very unbalanced on the left (with many low values and a few very high values).
+
 ### Nested means
 
-### Geometric progression
+The nested means method is used to create classes in a hierarchical fashion. Each class is defined by the average of the values of the parent class.
 
-This classification method allows you to create classes whose limits are defined by a geometric progression: each class is defined by a multiple of the previous one.
+The number of classes that can be chosen for this method is therefore necessarily a power of 2 (2, 4, 8, etc.).
 
 ### User-defined
 
 This method allows you to define the class limits manually.
+
+## Choosing a color progression
+
+Two types of color progression are available in Magrit:
+- Sequential palettes: used to represent continuously ordered data.
+- Divergent palettes: used to represent data ordered around a central value (e.g., an average, or the value zero, etc.).
+
+The available color palettes are taken from the [dicopal](https://github.com/riatelab/dicopal.js) library
+which offers palettes from a wide range of suppliers: ColorBrewer, Fabio Crameri's Scientific Colour Maps,
+CartoColors, CmOcean, Matplotlib, Light & Bartlein, MyCarta, Tableau, Joshua Stevens, etc.
+
+These palettes can be generated for any number of classes. For divergent palettes,
+options allow you to choose whether or not a central (neutral) class should be present, and
+the position of this class (or of the inflection point, if applicable), enabling you to generate asymmetrical divergent palettes.
+

docs/functionalities/classification.md

@@ -36,6 +36,7 @@ Cette méthode, parfois également appelée "amplitudes égales", permet de cré
 
 Cette méthode (notamment démocratisée par l'outil PhilCarto), permet d'effectuer une discrétisation selon la méthode des quartiles tout en isolant les valeurs extrêmes :
 elle produit ainsi 6 classes.
+Les 6 classes sont définies avec les bornes suivantes : minimum, percentile 5 (0.05%), 1er quartile (25%), médiane (50%), 3ème quartile (75%), percentile 95 (95%), maximum.
 
 ### Seuils naturels (algorithme CKMeans)
 
@@ -68,6 +69,8 @@ Cette méthode de discrétisation permet de créer des classes pour des séries
 
 ### Moyennes emboitées
 
+La méthode des moyennes emboitées permet de créer des classes de manière hiérarchique. Chaque classe est définie par la moyenne des valeurs de la classe parente (ou de l'ensemble de la série pour les deux premières classes).
+
 Le nombre de classes qu'il est possible de choisir pour cette méthode est donc nécessairement une puissance de 2 (2, 4, 8, etc.).
 
 ### Saisie manuelle