Celestial Body

Predicting Celestial Body using RandomForest

Installing libraries

library(tidyverse)

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.2     ✔ readr     2.1.4
✔ forcats   1.0.0     ✔ stringr   1.5.0
✔ ggplot2   3.4.2     ✔ tibble    3.2.1
✔ lubridate 1.9.2     ✔ tidyr     1.3.0
✔ purrr     1.0.1     
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(randomForest)

randomForest 4.7-1.1
Type rfNews() to see new features/changes/bug fixes.

Attaching package: 'randomForest'

The following object is masked from 'package:dplyr':

    combine

The following object is masked from 'package:ggplot2':

    margin

Importing training set as a dataframe

celestial_training <- read_csv("celestial_train.csv")

Rows: 50000 Columns: 18
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr  (1): class
dbl (17): id, alpha, delta, u, g, r, i, z, run_ID, rerun_ID, cam_col, field_...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

celestial_training

ABCDEFGHIJ0123456789

id <dbl>	alpha <dbl>	delta <dbl>	u <dbl>	g <dbl>	r <dbl>	i <dbl>
0	135.68910660	3.249463e+01	23.87882	22.27530	20.39501	19.16573
1	144.82610055	3.127418e+01	24.77759	22.83188	22.58444	21.16812
2	338.74103775	-4.028276e-01	22.13682	23.77656	21.61162	20.50454
3	340.99512051	2.058948e+01	23.48827	23.33776	21.32195	20.25615
4	200.29047539	4.719940e+01	24.40286	22.35669	20.61032	19.46490
5	39.14969060	2.810284e+01	21.74669	20.03493	19.17553	18.81823
6	328.09207617	1.822031e+01	25.77163	22.52042	20.63884	19.78071
7	331.50202998	1.003580e+01	20.82940	18.75091	17.51118	17.01631
8	344.98477027	-3.526158e-01	23.20911	22.79291	22.08589	21.86282
9	171.97542457	6.774745e+01	22.13367	20.84772	18.96537	18.31696

Making sure that the response variable is a factor

celestial_training <- celestial_training %>% 
  mutate(class = as.factor(class))

celestial_training %>% 
  select(class) %>% 
  distinct()

ABCDEFGHIJ0123456789

class <fct>
GALAXY
QSO
STAR

Creating model using randomForest

model.2 <- randomForest(class ~ .-(id+spec_obj_ID), data = celestial_training, ntree = 750, mtry = 4)

model.2


Call:
 randomForest(formula = class ~ . - (id + spec_obj_ID), data = celestial_training,      ntree = 750, mtry = 4) 
               Type of random forest: classification
                     Number of trees: 750
No. of variables tried at each split: 4

        OOB estimate of  error rate: 2.2%
Confusion matrix:
       GALAXY  QSO  STAR class.error
GALAXY  29314  341    74 0.013959434
QSO       682 8885     1 0.071383779
STAR        4    0 10699 0.000373727

Importing Testing Dataset

celestial_testing = read_csv("celestial_test.csv")

Rows: 50000 Columns: 17
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
dbl (17): id, alpha, delta, u, g, r, i, z, run_ID, rerun_ID, cam_col, field_...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Making predictions

testing_predictions.2 <- data.frame(predict(model.2, newdata = celestial_testing)) %>%
  rename("output" = "predict.model.2..newdata...celestial_testing.")

submission.2 <- data.frame(c(celestial_testing, testing_predictions.2)) %>%
  select(id, output)
submission.2

ABCDEFGHIJ0123456789

id <dbl>	output <fct>
50000	QSO
50001	GALAXY
50002	QSO
50003	GALAXY
50004	STAR
50005	GALAXY
50006	GALAXY
50007	QSO
50008	GALAXY
50009	GALAXY

Exporting predictions as csv

write.csv(submission.2, "submission_2.csv", row.names = FALSE)