sancheck 0.1.3

Minimal-tuning CLI tool for data stats and clarity checks

sancheck

What is this?

sancheck is a minimal-tuning CLI tool for quickly assessing the statistical cleanliness of numeric columns in CSV datasets.
It provides a fast, high-level overview before deeper analysis or modeling.

When should I use it?

• Before exploratory data analysis (EDA)
• Before training statistical or machine learning models
• When you want a quick sanity check without manual inspection

What it does NOT do

• It does not clean or modify data
• It does not model relationships
• It does not replace proper EDA or data validation pipelines

Quick start

Run the tool on a CSV file:

sancheck [csv_dir] [target_column_name] [n_feature_per_plot or 'all']

Example output

Summary of columns

• Valid numeric columns: 9
• Ignored non-numeric columns: 7

📌 Column problems

• Column with NaN/Inf/invalid: 0
  • Age: invalid=0/5000 (0.000)
  • Class: invalid=0/5000 (0.000)
  • Study_Hours_Per_Day: invalid=0/5000 (0.000)
  • Attendance_Percentage: invalid=0/5000 (0.000)
  • Math_Score: invalid=0/5000 (0.000)
  • Science_Score: invalid=0/5000 (0.000)
  • English_Score: invalid=0/5000 (0.000)
  • Previous_Year_Score: invalid=0/5000 (0.000)
  • Final_Percentage: invalid=0/5000 (0.000)
• Type inconsistency column: 0
  • Age: bad_type=0 (0.000)
  • Class: bad_type=0 (0.000)
  • Study_Hours_Per_Day: bad_type=0 (0.000)
  • Attendance_Percentage: bad_type=0 (0.000)
  • Math_Score: bad_type=0 (0.000)
  • Science_Score: bad_type=0 (0.000)
  • English_Score: bad_type=0 (0.000)
  • Previous_Year_Score: bad_type=0 (0.000)
  • Final_Percentage: bad_type=0 (0.000)
• Similar feature pairs: none

📌 Row problems

• Problematic rows (NaN/Inf): 0/5000
• Severity row: 0.132
  • row 4867: score=0.625, invalid=False
  • row 82: score=0.624, invalid=False
  • row 1364: score=0.622, invalid=False
  • row 1482: score=0.619, invalid=False
  • row 4913: score=0.611, invalid=False

📌 Distribution / interpretation

• High entropy means the distribution is more even/complex; it's not automatically 'noise', it can also be multimodal.

• High spread score means the data is more dispersed robustly compared to its central tendency.

  Top entropy:

  • Attendance_Percentage: entropy=1.000 (very spread / more uniform or complex distribution)
  • Science_Score: entropy=0.998 (very spread / more uniform or complex distribution)
  • English_Score: entropy=0.998 (very spread / more uniform or complex distribution)
  • Math_Score: entropy=0.997 (very spread / more uniform or complex distribution)
  • Study_Hours_Per_Day: entropy=0.997 (very spread / more uniform or complex distribution)

  Top spread:

  • Class: spread_score=0.690 (wide / large variation), var=1.225, iqr=1.000
  • Final_Percentage: spread_score=0.471 (moderate / moderate variation), var=120.211, iqr=15.660
  • Math_Score: spread_score=0.384 (moderate / moderate variation), var=350.606, iqr=32.000
  • Science_Score: spread_score=0.380 (moderate / moderate variation), var=366.385, iqr=33.000
  • Previous_Year_Score: spread_score=0.377 (moderate / moderate variation), var=261.065, iqr=28.000

📌 Structure

• VIF mean (normalized): 0.000
• VIF per-feature
  • Age: VIF=62.762
  • Class: VIF=58.647
  • Study_Hours_Per_Day: VIF=5.229
  • Attendance_Percentage: VIF=24.408
  • Math_Score: VIF=73695901.084
  • Science_Score: VIF=72203502.298
  • English_Score: VIF=74047280.603
  • Previous_Year_Score: VIF=17.149
  • Final_Percentage: VIF=627307333.939
• sparsity: 0.001
• class override ratio: 1.000

📌 Normality

• Shapiro-wilk and KS test score per-feature:
  • Age: Shapiro=0.000 | KS=0.000
  • Class: Shapiro=0.000 | KS=0.000
  • Study_Hours_Per_Day: Shapiro=0.000 | KS=0.000
  • Attendance_Percentage: Shapiro=0.000 | KS=0.000
  • Math_Score: Shapiro=0.000 | KS=0.000
  • Science_Score: Shapiro=0.000 | KS=0.000
  • English_Score: Shapiro=0.000 | KS=0.000
  • Previous_Year_Score: Shapiro=0.000 | KS=0.000
  • Final_Percentage: Shapiro=0.000 | KS=0.038
• normality score (based on skewness and kurtosis): 0.884

🧼 Cleanineess status

• cleanliness score: 0.961 / 1.000
• cleanliness label: very clean
• missing severity: 0.000
• type severity: 0.000
• similarity severity: 0.000
• row severity: 0.132

📊 Dataset-level distribution summary

• avg entropy: 0.979
• avg spread score: 0.420

Interpretation tips

• Higher clarity scores indicate cleaner numeric data

• Anomalous rows are ranked, not classified — use them for inspection

• Non-numeric columns are ignored by design

• This tool is best used as a fast pre-analysis step