plots changed

plots theme changed lollipop plot: threshold agrument added, closed #2 radar plot: labels changed, text_start_point and text_size arguments added
ModelOriented · Mar 28, 2019 · 8929887 · 8929887
1 parent 3debdc9
commit 8929887
Show file tree

Hide file tree

Showing 50 changed files with 92 additions and 578 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,7 +1,7 @@
 Package: EIX
 Title: EIX: Explain Interactions in XGBoost
        Structure mining for XGBoost and LightGBM models
-Version: 0.1
+Version: 0.2
 Authors@R: c(
   person("Ewelina", "Karbowiak", email = "ewelina.karbowiak12@gmail.com", role = c("aut", "cre")),
   person("Przemyslaw", "Biecek", email = "przemyslaw.biecek@gmail.com", role = c("aut","ths"))

diff --git a/NAMESPACE b/NAMESPACE
@@ -9,7 +9,7 @@ export(lollipop)
 export(waterfall)
 import(data.table)
 import(ggplot2)
-importFrom(DALEX,theme_mi2)
+importFrom(DALEX,theme_drwhy)
 importFrom(ggiraphExtra,coord_radar)
 importFrom(ggrepel,geom_label_repel)
 importFrom(ggrepel,geom_text_repel)

diff --git a/R/lollipop.R b/R/lollipop.R
@@ -13,6 +13,7 @@
 #' @import data.table
 #'
 #' @examples
+#' options(warn = -1)
 #' library("EIX")
 #' library("Matrix")
 #' sm <- sparse.model.matrix(left ~ . - 1,  data = HR_data)

diff --git a/R/plot_importance.R b/R/plot_importance.R
@@ -27,6 +27,8 @@
 #'               six measures of variables' or interactions' importance in the model.
 #'               If FALSE the plot containing two chosen measures
 #'               of variables' or interactions' importance in the model.
+#' @param text_start_point place, where the names of the particular feature start. Available for `radar=TRUE`. Range from 0 to 1. Default 0.5.
+#' @param text_size size of the text on the plot. Default 3.5.
 #' @param xmeasure measure on the x-axis.Available for `radar=FALSE`. Default "sumCover".
 #' @param ymeasure measure on the y-axis. Available for `radar=FALSE`. Default "sumGain".
 #' @param ... other parameters.
@@ -35,7 +37,7 @@
 #'
 #' @import ggplot2
 #' @import data.table
-#' @importFrom DALEX theme_mi2
+#' @importFrom DALEX theme_drwhy
 #' @importFrom ggrepel geom_label_repel
 #' @importFrom ggiraphExtra coord_radar
 #'
@@ -83,7 +85,7 @@
 #' @export
 
 
-plot.importance <- function(x, ...,  top = 10, radar = TRUE,
+plot.importance <- function(x, ...,  top = 10, radar = TRUE, text_start_point = 0.5, text_size=3.5,
                                  xmeasure = "sumCover", ymeasure = "sumGain"){
 
   Feature <- sumGain <- sumCover <- meanGain <- meanCover <-
@@ -97,7 +99,7 @@ plot.importance <- function(x, ...,  top = 10, radar = TRUE,
     ggplot(data.frame(x[1:top, ]),
            aes_string(x = xmeasure, y = ymeasure, label = "Feature")) +
       geom_point() +
-      scale_size() + geom_label_repel() + theme_mi2()
+      scale_size() + geom_label_repel() + theme_drwhy()
 
   }else{
     import <- as.data.table(x[1:top, ])
@@ -108,25 +110,34 @@ plot.importance <- function(x, ...,  top = 10, radar = TRUE,
                               meanCover = meanCover / max(import[, meanCover]),
                               mean5Gain = mean5Gain / max(import[, mean5Gain]),
                               frequency = frequency / max(import[, frequency]))]
+    data<-import[,Feature:= ifelse(nchar(import[,Feature])>20, gsub(":", ": :",import[,Feature]),Feature)]
 
     import$Feature <- factor(import$Feature, levels = import$Feature[order(import$sumGain, decreasing = TRUE)])
-    data_to_plot <- melt(import, id = 1, measures = 2:6)
+
+    #angles and hjust of labels
+    numberOfBars=nrow(import)
+    angle= 90-360*(row(import)[,1]-0.5)/numberOfBars
+
+    import$hjust<-ifelse( angle < -90, 1, 0)
+    import$angle<-ifelse(angle < -90, angle+180, angle)
+
+    data_to_plot <- melt(import, id = c(1,8,9), measures = 2:6, value.factor = FALSE)
+    data<-data_to_plot[,.(hjust=mean(hjust),angle=mean(angle)), by=Feature]
 
     ggplot(data.frame(data_to_plot),
            aes(x = Feature, y = value, colour = variable, group = variable)) +
       geom_line(size = 1.5) +
       geom_point(size = 2.5) +
-      theme_mi2() +
+      theme_drwhy() +
       theme(axis.title.x = element_blank(),
             axis.title.y = element_blank(),
-            #axis.text.x = element_text(angle=45),
             legend.position = "bottom",
-            #axis.text.y = element_blank(),
             panel.grid.major.y = element_line(colour = "gray68", linetype = "dashed", size = 0.4),
             axis.line = element_blank(),
-            plot.margin = margin(40, 40, 40, 40)) +
-      labs(fill = "Measures") +
-      coord_radar()+
-      scale_x_discrete(labels = lapply(strwrap(import[,Feature], width = 10, simplify = FALSE), paste, collapse="\n"))
+            axis.text.x=element_blank(),) +
+      labs(fill = "Measures")+
+      coord_radar() +
+      geom_text(data=data, aes(x=Feature, y= rep(text_start_point,top), label=lapply(strwrap(data[,Feature], width = 10, simplify = FALSE), paste, collapse="\n"), hjust=hjust), color="#371ea3", fontface="bold",alpha=0.6, size=text_size, angle= data$angle, inherit.aes = FALSE )
+
   }
 }
diff --git a/R/plot_interactions.R b/R/plot_interactions.R
@@ -14,7 +14,7 @@
 #'
 #' @import data.table
 #' @import ggplot2
-#' @importFrom DALEX theme_mi2
+#' @importFrom DALEX theme_drwhy
 #'
 #' @examples
 #' library("EIX")
@@ -65,11 +65,11 @@ plot.interactions <- function(x, ...) {
 
   ggplot(data.frame(x), aes(Child, Parent, sumGain)) +
     geom_tile(aes(fill = breaks)) +
-    theme_mi2() +
+    theme_drwhy() +
     theme(axis.text.x = element_text(hjust = 1, angle = 90),
           axis.text.y = element_text(hjust = 1, angle = 0)) +
     scale_fill_manual(name = "sumGain",
-                      values = c("#ffffff", "#ccccff", "#7f7fff", "#3232ff"),
+                      values = c("#f4f5ff", "#ccccff", "#7f7fff", "#3232ff"),
                       drop = FALSE,
                       breaks = levels(x$breaks),
                       labels = c("very low", "low", "medium", "high")) +

diff --git a/R/plot_lollipop.R b/R/plot_lollipop.R
@@ -7,18 +7,21 @@
 #'               and variables in the roots (horizontal label) will be displayed,
 #'               if "interactions" then labels for all interactions,
 #'               if "roots" then labels for all variables in the root.
+#' @param threshold  on the plot will occur only labels with Gain higher than `threshold` of the max Gain value in the model.
+#'                   The lower threshold, the more labels on the plot. Range from 0 to 1. Default 0.1.
 #' @param log  TRUE/FALSE logarithmic scale on the plot. Default TRUE.
 #' @param ... other parameters.
 #'
 #' @return a ggplot object
 #'
 #' @import data.table
 #' @import ggplot2
-#' @importFrom DALEX theme_mi2
+#' @importFrom DALEX theme_drwhy
 #' @importFrom ggrepel geom_label_repel
 #' @importFrom ggrepel geom_text_repel
 #'
 #' @examples
+#' options(warn = -1)
 #' library("EIX")
 #' library("Matrix")
 #' sm <- sparse.model.matrix(left ~ . - 1,  data = HR_data)
@@ -42,7 +45,7 @@
 #'
 #' @export
 
-plot.lollipop<-function(x, ..., labels = "topAll", log = TRUE){
+plot.lollipop<-function(x, ..., labels = "topAll", log = TRUE, threshold=0.1){
 
   Tree <- Quality <- depth <- Feature <- NULL
   nodes <- x[[1]]
@@ -57,24 +60,23 @@ plot.lollipop<-function(x, ..., labels = "topAll", log = TRUE){
     switch(labels,
            topAll = {
              p + geom_text_repel(data = data.frame(nodes),
-                                 aes(label = ifelse((interaction == TRUE) & (Quality > 0.1 * (max(nodes[, Quality]))), Feature, '')),
+                                 aes(label = ifelse((interaction == TRUE) & (Quality > threshold * (max(nodes[, Quality]))), Feature, '')),
                                  angle = 90, nudge_y = 0.05, direction  = "x", vjust = 0, segment.size = 0.2) +
                  geom_label_repel(data = data.frame(roots),
-                                aes(label = ifelse(Quality > 0.1 * (max(nodes[, Quality] )), Feature, '')))
+                                aes(label = ifelse(Quality > threshold * (max(nodes[, Quality] )), Feature, '')))
            },
            interactions = {
              p + geom_text_repel(data = data.frame(nodes),
-                                                 aes(label = ifelse(( interaction == TRUE ), Feature, '')),
+                                                 aes(label = ifelse(( interaction == TRUE ) & (Quality > threshold * (max(nodes[, Quality]))), Feature, '')),
                                                  angle = 90, nudge_y = 0.05, direction  = "x", vjust = 0, segment.size = 0.2 )
            },
            roots = {
-             p + geom_label_repel(data = data.frame(roots), aes(label = Feature))
+             p + geom_label_repel(data = data.frame(roots), aes(label = ifelse(Quality > threshold * (max(nodes[, Quality])), Feature, '')))
            })}
 
-  p <- p + theme_mi2() + ylab("Gain") +
+  p <- p + theme_drwhy()+ ylab("Gain") +
     scale_shape_discrete("Depth") +
     scale_colour_discrete("Depth") + if (log) {scale_x_continuous(trans = 'log10')}
-
   p
 }
 
diff --git a/README.md b/README.md
@@ -5,7 +5,7 @@ A set of tools to explain XGBoost and LightGBM models.
 
 Install from GitHub
 ```{r}
-devtools::install_github("ekarbowiak/EIX")
+devtools::install_github("ModelOriented/EIX")
 ```
 
 ## Cheatsheets

diff --git a/cheatsheets/EIX.pdf b/cheatsheets/EIX.pdf
diff --git a/cheatsheets/EIX.png b/cheatsheets/EIX.png
diff --git a/cheatsheets/EIX.pptx b/cheatsheets/EIX.pptx
diff --git a/docs/articles/EIX.html b/docs/articles/EIX.html
diff --git a/docs/articles/EIX_files/figure-html/unnamed-chunk-10-1.png b/docs/articles/EIX_files/figure-html/unnamed-chunk-10-1.png
diff --git a/docs/articles/EIX_files/figure-html/unnamed-chunk-6-1.png b/docs/articles/EIX_files/figure-html/unnamed-chunk-6-1.png
diff --git a/docs/articles/EIX_files/figure-html/unnamed-chunk-7-1.png b/docs/articles/EIX_files/figure-html/unnamed-chunk-7-1.png
diff --git a/docs/articles/EIX_files/figure-html/unnamed-chunk-8-1.png b/docs/articles/EIX_files/figure-html/unnamed-chunk-8-1.png
diff --git a/docs/articles/EIX_files/figure-html/unnamed-chunk-9-1.png b/docs/articles/EIX_files/figure-html/unnamed-chunk-9-1.png
diff --git a/docs/index.html b/docs/index.html
diff --git a/docs/reference/importance-1.png b/docs/reference/importance-1.png
diff --git a/docs/reference/importance-2.png b/docs/reference/importance-2.png
diff --git a/docs/reference/importance-3.png b/docs/reference/importance-3.png
diff --git a/docs/reference/importance-4.png b/docs/reference/importance-4.png
diff --git a/docs/reference/importance-5.png b/docs/reference/importance-5.png
diff --git a/docs/reference/importance-6.png b/docs/reference/importance-6.png
diff --git a/docs/reference/index.html b/docs/reference/index.html
diff --git a/docs/reference/interactions-1.png b/docs/reference/interactions-1.png
diff --git a/docs/reference/interactions-2.png b/docs/reference/interactions-2.png
diff --git a/docs/reference/interactions-3.png b/docs/reference/interactions-3.png
diff --git a/docs/reference/interactions-4.png b/docs/reference/interactions-4.png
diff --git a/docs/reference/lollipop-1.png b/docs/reference/lollipop-1.png
diff --git a/docs/reference/lollipop-2.png b/docs/reference/lollipop-2.png