From 106637956c7dfd35a1e003a51003256f9f4e473d Mon Sep 17 00:00:00 2001
From: Drew Giffin <drewgiffin97@gmail.com>
Date: Sun, 19 Oct 2025 17:17:41 -0400
Subject: [PATCH] Cleaned data is now kept separate from raw

---
 main.py | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/main.py b/main.py
index 782bd8d..9fe235b 100644
--- a/main.py
+++ b/main.py
@@ -10,10 +10,10 @@ def main():
     df = load_data()
     
     #preprocessing
-    preprocess_data(df)
+    df_clean = preprocess_data(df)
     
     #exploratory data analysis
-    draw_plots(df)
+    draw_plots(df_clean)
     
 def load_data():
     df = pd.read_csv(data_path, encoding="ascii", delimiter=",")
@@ -39,7 +39,8 @@ def clean_data(df):
     print(df.isnull().sum())
     print("\n")
     
-    df.dropna(inplace=True)
+    df.dropna(inplace=False)
+    return df
 
 def order_data_stress_level(df):
     df["Stress_Level"] = pd.Categorical(
@@ -77,7 +78,8 @@ def draw_plots(df):
     display_feature_boxplots(df)
 
 def preprocess_data(df):
-    clean_data(df)
-    order_data_stress_level(df)
+    df_clean = clean_data(df)
+    order_data_stress_level(df_clean)
+    return df_clean
 
 main()
\ No newline at end of file