analysis.py

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df = pd.read_csv("student_data.csv")

print(df.head())
print("\n----Basic Info----")
print(df.info())
print("\n----Summary Statistics---")
print(df.describe())
print("\n---- Missing Values----")
print(df.isnull().sum())
print("Total Students:", len(df))
print(df.columns)



duplicates = df.duplicated().sum()
print("\nDuplicate rows found:", duplicates)
df = df.drop_duplicates()

print("\nNegative values check:")
print((df[["Study hours", 'Attendance', 'Marks']] < 0).sum())

print("\nFill Missing values")
df['Study hours'] = df['Study hours'].fillna(df['Study hours'].mean())
df['Attendance'] = df["Attendance"].fillna(df['Attendance'].mean())
df['Marks'] = df['Marks'].fillna(df['Marks'].mean())

# Performance Category
def Categorize_marks(m):
    if m >= 75:
        return "High"
    elif m >= 50:
        return "Medium"
    else:
        return "Low"

df["performance_category"] = df["Marks"].apply(Categorize_marks)

print("\nPerformance category distribution:")
print(df["performance_category"].value_counts())


# Basic Analysis
print("Average Study hours:", df["Study hours"].mean())
print("Average Attendance:", df["Attendance"].mean())
print("Average Marks:", df["Marks"].mean())

print("\nGender Wise Count:")
print(df['Gender'].value_counts())

# Relationship analysis
print("\n---Correlation with marks---")
print(df.corr(numeric_only=True))

# visualization
numeric_df = df.select_dtypes(include=['int64', 'float64'])

# correlation heatmap
plt.figure(figsize=(6, 4))
sns.heatmap(numeric_df.corr(), annot=True, cmap="Blues")
plt.title("Correlation heatmap")
plt.show()

# Study Hours vs Marks (Scatter + Trendline)
plt.figure(figsize=(6, 4))
sns.regplot(x='Study hours', y='Marks', data=df)
plt.xlabel("Study Hours")
plt.ylabel("Marks")
plt.title("Study Hours vs Marks (Trendline)")
plt.show()

# Attendance vs Marks (Scatter + Trendline)
plt.figure(figsize=(6, 4))
sns.regplot(x='Attendance', y='Marks', data=df)
plt.xlabel("Attendance")
plt.ylabel("Marks")
plt.title("Attendance vs Marks (Trendline)")
plt.show()

# Boxplots
plt.figure(figsize=(6, 4))
sns.boxplot(x='Gender', y="Marks", data=df)
plt.title("Marks Distribution by Gender")
plt.xlabel("Gender")
plt.ylabel("Marks")
plt.show()

plt.figure(figsize=(6, 4))
sns.boxplot(x='performance_category', y="Marks", data=df)
plt.title("Marks Distribution by Performance Category")
plt.xlabel("Performance Category")
plt.ylabel("Marks")
plt.show()

# ---- Histograms & Distribution Plots ----

plt.figure(figsize=(6,4))
sns.histplot(df["Marks"], kde=True)
plt.title("Marks Distribution")
plt.xlabel("Marks")
plt.ylabel("Count")
plt.show()

plt.figure(figsize=(6,4))
sns.histplot(df["Study hours"], kde=True)
plt.title("Study Hours Distribution")
plt.xlabel("Study Hours")
plt.ylabel("Count")
plt.show()

plt.figure(figsize=(6,4))
sns.histplot(df["Attendance"], kde=True)
plt.title("Attendance Distribution")
plt.xlabel("Attendance")
plt.ylabel("Count")
plt.show()