In [12]:
library(tidyverse)
library(MLmetrics)
In [13]:
pandas = read_tsv("~/RIDIR/Datasets/AreaInterpolateValidation/geopandas_intensive2000.tsv", col_names = c("TID", "pandas")) %>%
    arrange(desc(pandas))
head(pandas)
nrow(pandas)

Parsed with column specification:
cols(
  TID = col_character(),
  pandas = col_double()
)
TIDpandas
T4210100032438.12638
T4210100015921.81991
T4210100016021.62135
T4210100033020.78395
T4210100022920.62140
T4210100018420.56308
367
In [14]:
spark = read_tsv("~/RIDIR/Datasets/AreaInterpolateValidation/geospark_intensive2000.tsv", col_names = c("TID", "spark")) %>%
    arrange(desc(spark))
head(spark)
nrow(spark)

Parsed with column specification:
cols(
  TID = col_character(),
  spark = col_double()
)
TIDspark
T4210100032438.12638
T4210100015921.81991
T4210100016021.62135
T4210100033020.78395
T4210100022920.62140
T4210100018420.56308
367
In [15]:
table = pandas %>% left_join(spark, by = c("TID"))
head(table)
nrow(table)
TIDpandasspark
T4210100032438.12638 38.12638
T4210100015921.81991 21.81991
T4210100016021.62135 21.62135
T4210100033020.78395 20.78395
T4210100022920.62140 20.62140
T4210100018420.56308 20.56308
367
In [16]:
head(table %>% filter(is.na(spark)))
TIDpandasspark
In [17]:
table = spark %>% inner_join(pandas, by = c("TID")) %>% 
    mutate(diff = abs(pandas - spark)) %>%
    arrange(desc(diff))
head(table)
nrow(table)
TIDsparkpandasdiff
T4210100032438.12638 38.12638 7.105427e-15
T4210100015921.81991 21.81991 7.105427e-15
T4210100033020.78395 20.78395 7.105427e-15
T4210100036318.76598 18.76598 7.105427e-15
T4210100033118.11994 18.11994 7.105427e-15
T4210100029814.91979 14.91979 5.329071e-15
367
In [18]:
ggplot(data = table, aes(x = pandas, y = spark)) + geom_point()
In [19]:
print(paste("R2:   ", R2_Score(table$spark, table$pandas)))
print(paste("MAE:  ",      MAE(table$spark, table$pandas)))
print(paste("MSE:  ",      MSE(table$spark, table$pandas)))
print(paste("RMSE: ",     RMSE(table$spark, table$pandas)))
print(paste("RAE:  ",      RAE(table$spark, table$pandas)))

[1] "R2:    1"
[1] "MAE:   5.67037145364048e-16"
[1] "MSE:   1.81290538282176e-30"
[1] "RMSE:  1.34644174876664e-15"
[1] "RAE:   1.00237609882984e-16"
In [ ]: