app/indicators.py

import numpy as np
import pandas as pd


def add_technical_indicators(df: pd.DataFrame) -> pd.DataFrame:
    df = df.copy()

    # RSI
    delta = df["Close"].diff()
    gain = delta.where(delta > 0, 0).rolling(window=14).mean()
    loss = -delta.where(delta < 0, 0).rolling(window=14).mean()
    rs = gain / loss
    df["RSI"] = 100 - (100 / (1 + rs))

    # MACD
    exp1 = df["Close"].ewm(span=12, adjust=False).mean()
    exp2 = df["Close"].ewm(span=26, adjust=False).mean()
    df["MACD"] = exp1 - exp2

    # Simple 50-day MA
    df["MA50"] = df["Close"].rolling(window=50).mean()

    # Log Volume
    df["Log_Volume"] = np.log(df["Volume"] + 1)

    df = df.dropna()
    return df


def classify_signal(predicted_change: float, asset_type: str = "stock") -> str:
    """
    predicted_change: percentage change predicted by model
    asset_type: 'stock' or 'crypto'
    """

    # Crypto → larger natural volatility
    if asset_type == "crypto":
        if predicted_change > 3.0:
            return "STRONG BUY"
        if predicted_change > 0.8:
            return "BUY"
        if predicted_change < -3.0:
            return "STRONG SELL"
        if predicted_change < -0.8:
            return "SELL"
        return "HOLD"

    # Stock → tighter thresholds
    if predicted_change > 1.5:
        return "STRONG BUY"
    if predicted_change > 0.3:
        return "BUY"
    if predicted_change < -1.5:
        return "STRONG SELL"
    if predicted_change < -0.3:
        return "SELL"
    return "HOLD"


def classify_volatility(df: pd.DataFrame, asset_type: str = "stock") -> str:
    returns = df["Close"].pct_change().dropna()
    vol = returns.std()

    # Crypto volatility bands
    if asset_type == "crypto":
        if vol < 0.02:
            return "Low"
        if vol < 0.05:
            return "Medium"
        return "High"

    # Stock volatility bands
    if vol < 0.008:
        return "Low"
    if vol < 0.02:
        return "Medium"
    return "High"