From 4da5529cae52a7130619cd691a42db5802016a98 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Andrzej=20W=C3=B3jtowicz?= <andre@amu.edu.pl>
Date: Sat, 16 Apr 2016 13:57:54 +0200
Subject: [PATCH] Added features statistics and raw numbers of class imbalance

---
 README.md       | 94 +++++++++++++++++++++++++++++++++++++++++++------
 readme-make.Rmd | 14 +++++++-
 2 files changed, 96 insertions(+), 12 deletions(-)

diff --git a/README.md b/README.md
index 840a993..9657b7e 100644
--- a/README.md
+++ b/README.md
@@ -3,7 +3,7 @@ Andrzej Wójtowicz
 
 
 
-Document generation date: 2016-04-16 01:21:55.
+Document generation date: 2016-04-16 13:55:00.
 
 
 
@@ -62,7 +62,16 @@ Document generation date: 2016-04-16 01:21:55.
 
 ```
 
-**Class imbalance**: 11% / 89%
+**Predictors**:
+
+|Class          | Frequency|
+|:--------------|---------:|
+|factor         |         6|
+|integer        |         3|
+|numeric        |         5|
+|ordered factor |         3|
+
+**Class imbalance**: 11% / 89% (4254 / 33973)
 
 ---
 
@@ -121,7 +130,13 @@ https://archive.ics.uci.edu/ml/citation_policy.html
 
 ```
 
-**Class imbalance**: 37% / 63%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|numeric |        30|
+
+**Class imbalance**: 37% / 63% (212 / 357)
 
 ---
 
@@ -158,7 +173,13 @@ O. L. Mangasarian and W. H. Wolberg: "Cancer diagnosis via linear programming",
 
 ```
 
-**Class imbalance**: 35% / 65%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|integer |         9|
+
+**Class imbalance**: 35% / 65% (239 / 444)
 
 ---
 
@@ -214,7 +235,16 @@ Ayres de Campos et al. (2000) SisPorto 2.0 A Program for Automated Analysis of C
 
 ```
 
-**Class imbalance**: 22% / 78%
+**Predictors**:
+
+|Class          | Frequency|
+|:--------------|---------:|
+|factor         |         9|
+|integer        |        17|
+|numeric        |         2|
+|ordered factor |         1|
+
+**Class imbalance**: 22% / 78% (471 / 1655)
 
 ---
 
@@ -264,7 +294,14 @@ Yeh, I. C., & Lien, C. H. (2009). The comparisons of data mining techniques for
 
 ```
 
-**Class imbalance**: 22% / 78%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|factor  |         3|
+|integer |        20|
+
+**Class imbalance**: 22% / 78% (6636 / 23364)
 
 ---
 
@@ -302,7 +339,15 @@ https://archive.ics.uci.edu/ml/citation_policy.html
 
 ```
 
-**Class imbalance**: 29% / 71%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|factor  |         1|
+|integer |         4|
+|numeric |         5|
+
+**Class imbalance**: 29% / 71% (167 / 416)
 
 ---
 
@@ -341,7 +386,13 @@ https://archive.ics.uci.edu/ml/citation_policy.html
 
 ```
 
-**Class imbalance**: 35% / 65%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|numeric |        10|
+
+**Class imbalance**: 35% / 65% (6688 / 12332)
 
 ---
 
@@ -383,7 +434,14 @@ Sikora M., Wrobel L.: Application of rule induction algorithms for analysis of d
 
 ```
 
-**Class imbalance**: 7% / 93%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|factor  |         4|
+|integer |        11|
+
+**Class imbalance**: 7% / 93% (170 / 2414)
 
 ---
 
@@ -470,7 +528,14 @@ https://archive.ics.uci.edu/ml/citation_policy.html
 
 ```
 
-**Class imbalance**: 39% / 61%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|integer |         2|
+|numeric |        55|
+
+**Class imbalance**: 39% / 61% (1813 / 2788)
 
 ---
 
@@ -511,7 +576,14 @@ P. Cortez, A. Cerdeira, F. Almeida, T. Matos and J. Reis. Modeling wine preferen
 
 ```
 
-**Class imbalance**: 37% / 63%
+**Predictors**:
+
+|Class   | Frequency|
+|:-------|---------:|
+|factor  |         1|
+|numeric |        11|
+
+**Class imbalance**: 37% / 63% (2384 / 4113)
 
 ---
 
diff --git a/readme-make.Rmd b/readme-make.Rmd
index 86970c2..b5496c0 100644
--- a/readme-make.Rmd
+++ b/readme-make.Rmd
@@ -64,11 +64,23 @@ for (dir.name in dir(PATH_DATASETS))
     cat(str(dataset))
     cat("\n```\n\n")
     
+    cat("**Predictors**:\n\n")
+    
+    df.pred = data.frame(table(sapply(dataset[, 1:(ncol(dataset)-1)], 
+                               function(f){paste(class(f), collapse=" ")})))
+    colnames(df.pred) = c("Class", "Frequency")
+    
+    cat(knitr::kable(df.pred, format="markdown"), sep="\n")
+    cat("\n")
+    
     perc.classes = sort(round(100*as.numeric(
         table(dataset[, ncol(dataset)]))/nrow(dataset), 0))
+    num.classes = sort(as.numeric(table(dataset[, ncol(dataset)])))
     cat(paste("**Class imbalance**:", 
               paste0(perc.classes[1], "% / ", 
-                     perc.classes[2], "%\n\n")))
+                     perc.classes[2], "% (",
+                     num.classes[1], " / ", 
+                     num.classes[2], ")\n\n")))
     cat("---\n\n")
 }
 ```