
 道  michi inspect  ·  customers.csv

  600 rows × 14 columns  ·  14.5% of cells missing  ·  0 duplicate rows  ·      
  target purchased                                                              
  sha256 0dbfe06c8719  ·  42.0 KB                                               
                                                                                
  column         kind          missing   unique   summary                       
  ──────────────────────────────────────────────────────────────────────────────
  age            numeric             —       45   mean 42.702  ·  range 20–64   
  salary         numeric         13.5%      511   mean 39,357.938  ·  range     
                                                  19,798–56,542  ·  1 outliers  
  region         categorical         —        4   top: north (239), south (188),
                                                  east (117)                    
  cabin          categorical     89.0%       66   top: C13 (1), C37 (1), C38 (1)
  notes          empty          100.0%        0                                 
  country        categorical         —        1   top: JP (600)                 
  country_copy   categorical         —        1   top: JP (600)                 
  record_id      text                —      600   length 7–7                    
  fare           numeric             —      467   mean 210.883  ·  range        
                                                  5.29–10,000  ·  skew +6.87  · 
                                                  37 outliers                   
  amount_text    text                —      598   length 5–6                    
  signup_date    categorical         —       11   top: 2024-01-15 (55),         
                                                  2024-02-15 (55), 2024-03-15   
                                                  (55)                          
  age_months     numeric             —       45   mean 512.42  ·  range 240–768 
  outcome_code   categorical         —        2   top: class-0 (363), class-1   
                                                  (237)                         
  purchased      numeric             —        2   mean 0.395  ·  range 0–1      
                                                                                
  Findings (14)                                                                 
                                                                                
  high    country                only one distinct value (JP)                   
  high    country_copy           only one distinct value (JP)                   
  high    notes                  every value is missing                         
  high    cabin                  89.0% missing (534 of 600)                     
  high    outcome_code           each of its 2 values maps to exactly one       
                                 'purchased' class                              
  warn    signup_date            values parse as dates but are stored as text   
  warn    country, country_copy  identical values in country, country_copy      
  warn    cabin                  66 distinct values across 66 rows              
  warn    fare                   skew 6.87 (right-tailed)                       
  warn    age, age_months        correlation +1.000                             
  warn    salary                 13.5% missing (81 of 600)                      
  warn    amount_text            100% of values parse as numbers                
  warn    fare                   37 values beyond 1.5×IQR (6.2%)                
  info    record_id              every value is distinct (looks like an         
                                 identifier)                                    
                                                                                
  Run again with --explain for what each finding means and your options.        
                                                                                
