presmerats/House-price-prediction

Framework description

Files

• main.R
• aux_functions.R
• 01_datapreparation_david.R
• linear_regression_fitting.R
• unsupervised_analysis.R
• Cross_validation.R
• trees_methods_fitting.R
• .. and more ..

Discarded files? (this codes are copied to main.R)

• execution_control.R
• main_datasets_generator.R
• main_experiments.R
• main_unsupervised_analysis.R

---

Results storage

• ./Dataset

  • preprocessed data,
    • clean, outliers, missing, pca, log-transforms, subsets

• ./Analysis Results

  • model_results.csv

    • for each model and dataset training error, testing error
    • written from inside the model function, or the CV function

  • MLproject.xslx

    • overview of the models and datasets executed

  • /Linear Model Fitting

    • images or other evidence or results
    • txt explaining main results and next steps

  • /Unsupervised Analysis

    • images or other evidence or results
    • 
    • txt explaining main results and next steps

  • for other analyses and models the same idea

---

main.R

rm(list = ls())
gc()


#### 1 - Set up environment ####

a = Sys.info()

david_wd <- "~/MIRI-Data Science/ML - Machine Learning/Project/mlproject/src"
asaf_wd <- ""
pau_wd <- "~/Projectes/mlproject/src"

if (a["user"] == "david") {
  wd <- david_wd
  } else if (a["user"] == "pau") {
   wd <- pau_wd
} else {
  wd <- asaf_wd
}
setwd(wd)

# source scripts

source("aux_functions.R")

source_scripts()

load_packages()





#### 2 - Read data ####

if (a[1] == "Linux"){
  # Linux reading file
  data <- read.csv(file="../Dataset/kc_house_data.csv", header=TRUE, sep=",")
  
} else {
  # windows reading file
  data <- read.csv(file="../Dataset/kc_house_data.csv", header=TRUE, sep=",")
}

#### 2.1 - Create datasets to be used ####

raw_continuous_dataset <- raw_continuous_vars_selection(data)
save(file = "../Dataset/raw_continuous_dataset.Rda", raw_continuous_dataset)
rm(raw_continuous_dataset)

#### 3 - Unsupervised analysis of the data ####

# test normality. 

# test-normality()

# perform PCA

pca_analysis()
pca_analysis_2()


# perform clustering





#### 4 - Linear Models fitting ####

load(file="../Dataset/raw_continuous_dataset.Rda")
linear_regression_fitting(raw_continuous_dataset, dataset_id = "raw_continuous_vars")

---