merge: Merge feat/analysis into dev

inspect_data.py

@@ -0,0 +1,59 @@
+import pandas as pd
+import pickle
+
+def inspect_and_generate_schemas():
+    # Load Data
+    print("Loading data...")
+    df = pd.read_csv('01_merged_data.csv')
+    
+    # Define features explicitly based on analysis
+    excluded_cols = [
+        'id_employee', 'eval_number', 'code_sondage', 
+        'a_quitte_l_entreprise', 'turnover'
+    ]
+    
+    features = [c for c in df.columns if c not in excluded_cols]
+    print(f"Selected features: {features}")
+
+    # Generate Pydantic Schema
+    pydantic_fields = []
+    sql_columns = []
+    
+    type_mapping = {
+        'int64': 'int',
+        'float64': 'float',
+        'object': 'str',
+        'bool': 'bool'
+    }
+    
+    sql_type_mapping = {
+        'int64': 'Integer',
+        'float64': 'Float',
+        'object': 'String',
+        'bool': 'Boolean'
+    }
+
+    for feature in features:
+        dtype = str(df[feature].dtype)
+        py_type = type_mapping.get(dtype, 'str')
+        sql_type = sql_type_mapping.get(dtype, 'String')
+        
+        pydantic_fields.append(f"    {feature}: {py_type}")
+        sql_columns.append(f"    {feature} = Column({sql_type})")
+
+    pydantic_schema = "class InputSchema(BaseModel):\n" + "\n".join(pydantic_fields)
+    
+    print("\n--- Generated Pydantic Schema ---")
+    print(pydantic_schema)
+    
+    print("\n--- Generated SQL Columns (for reference) ---")
+    print("\n".join(sql_columns))
+    
+    # Save to file
+    with open('generated_schemas_v2.txt', 'w') as f:
+        f.write(pydantic_schema)
+        f.write("\n\n")
+        f.write("\n".join(sql_columns))
+
+if __name__ == "__main__":
+    inspect_and_generate_schemas()