import pandas as pd
import numpy as np


# this is a dictionary i created for the model like a fake model...

ehr_fields = {
    "patient_id": lambda: pd.Series(range(10000, 10100)),
    "age": lambda: pd.Series([round(x) for x in np.random.uniform(18, 90, size=100)]),
    "gender": lambda: pd.Series(np.random.choice(["M", "F", "O"], size=100)),
    "race": lambda: pd.Series(np.random.choice(["White", "Black", "Asian", "Hispanic"], size=100)),
    "ethnicity": lambda: pd.Series(np.random.choice(["Non-Hispanic", "Hispanic"], size=100)),
    "diagnosis": lambda: pd.Series(np.random.choice([
        "Hypertension", "Diabetes", "Asthma", "Heart Failure", "Obesity"
    ], size=100)),
    "medication": lambda: pd.Series(np.random.choice([
        "Metformin", "Lisinopril", "Albuterol", "Insulin", "Atorvastatin"
    ], size=100)),
    "visit_duration": lambda: pd.Series(np.random.randint(5, 180, size=100)),
    "readmitted": lambda: pd.Series(np.random.choice(["Yes", "No"], size=100))
}


def generate_synthetic_ehr(num_records=100):
    """Generate synthetic EHR data based on schema"""
    data = {}
    for field, generator in ehr_fields.items():
        data[field] = generator()

    df = pd.DataFrame(data)
    return df.head(num_records)