src/analyze/data_report.py

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

# Load dataset
df = pd.read_csv("data/processed/text_all_clean.csv")

# Basic info
print("🔹 Dataset Shape:", df.shape)
print("\n🔹 Columns:", df.columns.tolist())

# Check missing values
print("\n🔹 Missing Values:\n", df.isnull().sum())

# Label distribution
print("\n🔹 Label Distribution:\n", df['label'].value_counts())

# Percentage distribution
label_percent = df['label'].value_counts(normalize=True) * 100
print("\n🔹 Label Percentage:\n", label_percent)

# Plot distribution
plt.figure()
sns.countplot(x='label', data=df)
plt.title("Label Distribution")
plt.xlabel("Class (0 = Non-suicidal, 1 = Suicidal)")
plt.ylabel("Count")
plt.show()

# Optional: language distribution (if useful)
if 'lang' in df.columns:
    print("\n🔹 Language Distribution:\n", df['lang'].value_counts())

# Check text length stats
df['text_length'] = df['text'].astype(str).apply(len)

print("\n🔹 Text Length Stats:\n", df['text_length'].describe())

plt.figure()
sns.histplot(df['text_length'], bins=50)
plt.title("Text Length Distribution")
plt.show()