data.py

import requests
import pandas as pd
import numpy as np
from datetime import datetime
import time
from ta.momentum import RSIIndicator
from ta.trend import MACD, EMAIndicator
from ta.volatility import BollingerBands

# --- KONFIGURASI ---
SYMBOL = 'BTCUSDT'
INTERVAL = '15m'
START_DATE = '01 Jan 2023'  # Kita ambil data dari awal 2023 biar model pintar
FILENAME = 'bitcoin_15m_lengkap.csv'

def get_binance_data(symbol, interval, start_str):
    print(f"🚀 Mulai menambang data {symbol} timeframe {interval} dari {start_str}...")
    
    api_url = "https://api.binance.com/api/v3/klines"
    start_ts = int(datetime.strptime(start_str, '%d %b %Y').timestamp() * 1000)
    end_ts = int(time.time() * 1000)
    
    data = []
    
    while start_ts < end_ts:
        params = {
            'symbol': symbol,
            'interval': interval,
            'limit': 1000,
            'startTime': start_ts
        }
        
        try:
            response = requests.get(api_url, params=params)
            klines = response.json()
            
            if not klines:
                break
                
            data.extend(klines)
            start_ts = klines[-1][0] + 1
            
            # Progress bar manual biar gak bosen nunggu
            last_date = datetime.fromtimestamp(klines[-1][0]/1000).strftime('%Y-%m-%d')
            print(f"   Sedot data sampai: {last_date} | Total baris: {len(data)}")
            
            time.sleep(0.5) # Biar gak kena banned Binance (sopan dikit)
            
        except Exception as e:
            print(f"❌ Error: {e}")
            break
            
    # Convert ke DataFrame rapi
    df = pd.DataFrame(data, columns=[
        'timestamp', 'open', 'high', 'low', 'close', 'volume', 
        'close_time', 'quote_asset_volume', 'trades', 
        'taker_buy_base', 'taker_buy_quote', 'ignore'
    ])
    
    # Ubah tipe data string jadi angka (float)
    numeric_cols = ['open', 'high', 'low', 'close', 'volume', 'trades']
    df[numeric_cols] = df[numeric_cols].astype(float)
    
    # Rapikan waktu
    df['datetime'] = pd.to_datetime(df['timestamp'], unit='ms')
    df = df[['datetime', 'open', 'high', 'low', 'close', 'volume', 'trades']]
    
    return df

def add_indicators(df):
    print("🛠️  Sedang meracik indikator teknikal...")
    
    # 1. RSI (Momentum)
    df['RSI'] = RSIIndicator(close=df['close'], window=14).rsi()
    
    # 2. MACD (Trend)
    macd = MACD(close=df['close'])
    df['MACD'] = macd.macd()
    df['MACD_signal'] = macd.macd_signal()
    
    # 3. EMA (Trend Halus)
    df['EMA_50'] = EMAIndicator(close=df['close'], window=50).ema_indicator()
    df['EMA_200'] = EMAIndicator(close=df['close'], window=200).ema_indicator()
    
    # 4. Bollinger Bands (Volatilitas)
    bb = BollingerBands(close=df['close'], window=20, window_dev=2)
    df['BB_Upper'] = bb.bollinger_hband()
    df['BB_Lower'] = bb.bollinger_lband()
    
    # 5. Target Label (Yang mau ditebak)
    # 1 jika Close candle BERIKUTNYA lebih tinggi dari Close sekarang
    df['Target_Next_Close'] = df['close'].shift(-1)
    df['Label'] = (df['Target_Next_Close'] > df['close']).astype(int)
    
    # Hapus baris yang ada NaN (kosong karena perhitungan indikator di awal data)
    df.dropna(inplace=True)
    return df

# --- EKSEKUSI ---
df = get_binance_data(SYMBOL, INTERVAL, START_DATE)
df = add_indicators(df)

print(f"✅ Selesai! Data bersih tersimpan: {len(df)} baris")
df.to_csv(FILENAME, index=False)
print(f"📁 File disimpan sebagai: {FILENAME}")