import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import train_test_split
from sklearn.metrics import r2_score
import joblib

# Generate synthetic training data for Hemoglobin model
np.random.seed(42)
size = 200
data = {
    "mean_intensity": np.random.uniform(0.2, 0.5, size),
    "bbox_width": np.random.uniform(0.05, 0.2, size),
    "bbox_height": np.random.uniform(0.05, 0.2, size),
    "eye_dist": np.random.uniform(0.2, 0.5, size),
    "nose_len": np.random.uniform(0.2, 0.5, size),
    "jaw_width": np.random.uniform(0.2, 0.5, size),
    "avg_skin_tone": np.random.uniform(0.2, 0.5, size),
    "hemoglobin": np.random.uniform(10.5, 17.5, size)  # realistic Hb range
}
df = pd.DataFrame(data)

# Save dataset
df.to_csv("hemoglobin_dataset.csv", index=False)

# Train-test split
X = df.drop(columns=["hemoglobin"])
y = df["hemoglobin"]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Train model
model = RandomForestRegressor(n_estimators=100, random_state=42)
model.fit(X_train, y_train)

# Evaluate
y_pred = model.predict(X_test)
print("R2 Score:", r2_score(y_test, y_pred))

# Save model
joblib.dump(model, "hemoglobin_model.pkl")
print("Model saved as hemoglobin_model.pkl")