Resultados clasificación

library(readr)
dataset <- read_csv("classification-results.csv")
## Parsed with column specification:
## cols(
##   .default = col_double(),
##   dataset = col_character(),
##   var = col_character(),
##   config.train = col_character(),
##   config.vars = col_character(),
##   T = col_integer(),
##   alg = col_character()
## )
## See spec(...) for full column specifications.
library(ggplot2)
library(dplyr)
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
library(tidyr)

dataset[,8:ncol(dataset)] <- round(dataset[,8:ncol(dataset)],digits = 3)

Creación de dataset genérico para resultados clasificación

df <- dataset %>%
  unite(dataset, 
        col=label,c("dataset","var","config.train","config.vars","T","alg"),
        sep = "-",remove=FALSE) %>% 
  select(label,dataset,var,config.train,config.vars,T,alg,FAR,Recall,Specificity,Accuracy,Kappa,F1,Precision)

Valores posibles en la columna vars Son las estaciones meteorológicas

print(unique(dataset$var))
## [1] "junin.temp_min"       "tunuyan.temp_min"     "las_paredes.temp_min"
## [4] "la_llave.temp_min"    "agua_amarga.temp_min"

Caso columna o dato que evalue si colabora o no el enfoque agregar vecinos.

El dataset del que hacemos la resta es la siguiente:

df %>%
  select(dataset,var,config.train,config.vars,T,alg,FAR) %>%
  filter(dataset == "dacc" & var == "junin.temp_min" ) %>% # & alg=="glm" & T == 1 ) %>%
  group_by(dataset,var,config.train,T,alg) %>% 
  arrange(dataset,var,config.train,T,alg) %>% 
  mutate(local_vs_all = lag(FAR) - FAR)
## # A tibble: 40 x 8
## # Groups:   dataset, var, config.train, T, alg [20]
##    dataset var     config.train config.vars     T alg     FAR local_vs_all
##    <chr>   <chr>   <chr>        <chr>       <int> <chr> <dbl>        <dbl>
##  1 dacc    junin.… normal       all             1 C5.0   0.23       NA    
##  2 dacc    junin.… normal       local           1 C5.0   0.2         0.03 
##  3 dacc    junin.… normal       all             1 glm    0.04       NA    
##  4 dacc    junin.… normal       local           1 glm    0.15       -0.110
##  5 dacc    junin.… normal       all             1 rf     0.16       NA    
##  6 dacc    junin.… normal       local           1 rf     0.15        0.01 
##  7 dacc    junin.… normal       all             1 rpart  0          NA    
##  8 dacc    junin.… normal       local           1 rpart  0.18       -0.18 
##  9 dacc    junin.… normal       all             2 C5.0   0.28       NA    
## 10 dacc    junin.… normal       local           2 C5.0   0.21        0.07 
## # ... with 30 more rows

aplico los filtros que quiera graficar/consultar y el orden

df1 <- df %>%
  select(dataset,var,config.train,config.vars,T,alg,FAR) %>%
  filter(dataset == "dacc" & var == "junin.temp_min" ) %>% # & alg=="glm" & T == 1 ) %>%
  group_by(dataset,var,config.train,T,alg) %>% 
  arrange(dataset,var,config.train,T,alg) %>% 
  mutate(local_vs_all = lag(FAR) - FAR) %>%
  filter(!is.na(local_vs_all))  %>%
  unite(col=label,c("dataset","var","config.train","T","alg"),
        sep = "-",remove=FALSE) 

ggplot(data=df1, aes(x= reorder(label,-local_vs_all), y=local_vs_all)) +
  geom_bar(stat="identity",fill="green")+ 
  geom_text(aes(label=local_vs_all), vjust=1.3, color="black", size=3) + 
  coord_flip()+
  theme_minimal() +
  labs(x = "Models", y = "local_vs_all_segun_FAR")

MOSTRAR RANKING POR recall, precision, F1, sensitivity, etc

df2 <- df %>%
  filter(dataset == "dacc" & var == "junin.temp_min" )  %>% # & alg=="glm" & T == 1 ) %>%
  arrange(FAR)

Donde dice FAR se puede cambiar por: Recall,Specificity,Accuracy,Kappa,F1,Precision

ggplot(data=df2, aes(x= reorder(label,-FAR), y=FAR, color=config.vars)) +
  geom_bar(stat="identity",fill="white")+ 
  geom_text(aes(label=FAR), vjust=1.3, color="black", size=3) + 
  coord_flip()+
  theme_minimal() +
  labs(x = "Models", y = "FAR", title="FAR")

ggplot(data=df2, aes(x= reorder(label,Recall), y=Recall, color=config.vars)) +
  geom_bar(stat="identity",fill="white")+ 
  geom_text(aes(label=Recall), vjust=1.3, color="black", size=3) + 
  coord_flip()+
  theme_minimal() +
  labs(x = "Models", y = "Recall", title="Recall")

Comparativas de sensitivity, precision, F1,

df3 <- df %>%
  select(label,dataset,var,config.train,config.vars,T,alg,Recall,Specificity,Accuracy,Kappa,F1,Precision) %>%
  filter(dataset == "dacc" & var == "junin.temp_min"  & alg=="glm") # & T == 1 ) %>%

library(reshape2)
## 
## Attaching package: 'reshape2'
## The following object is masked from 'package:tidyr':
## 
##     smiths
df3 <- melt(df3,id=(c("label","dataset","var","config.train","config.vars","T","alg")))

  
ggplot(data=df3, aes(x= reorder(label,value), y=value, fill=variable)) +
geom_bar(stat="identity",position=position_dodge())+ 
coord_flip()+
theme(legend.position="bottom") +
labs(x = "Logistic Regression Models", y = "metricas", title="comparacion")