1) Data — Collect & Save OHLC (MT5)

This notebook fetches OHLCV data from MT5 via a tiny adapter and saves it to CSV.

Keep your credentials in .env and do not commit .env.
If MT5 is not available, you can skip this and drop in a CSV manually.

In [1]:

import sys, os; sys.path.append(os.path.abspath('..')) # Path fix

# Prereqs: `pip install -r requirements.txt` (locally)
import os
from pathlib import Path
import pandas as pd
from dotenv import load_dotenv

from adapters import broker

# Load settings from env
load_dotenv()
SYMBOL = os.getenv("TRAINING_SYMBOL", "EURUSD")
TIMEFRAME = os.getenv("TIMEFRAME", "M15")
START = os.getenv("DATA_START", "2023-01-01")
END   = os.getenv("DATA_END", "2025-01-01")

print("Connecting to MT5...")
ok = broker.open_session(
    login=int(os.getenv("MT5_LOGIN", "0")) or None,
    password=os.getenv("MT5_PASSWORD"),
    server=os.getenv("MT5_SERVER")
)
print("MT5 connected:", ok)

Connecting to MT5...
MT5 connected: True

In [2]:


# Fetch OHLC and save
DATA_DIR = Path("data"); DATA_DIR.mkdir(exist_ok=True)
df = broker.fetch_ohlc(SYMBOL, TIMEFRAME, START, END)
out_csv = DATA_DIR / f"ohlc_{SYMBOL}_{TIMEFRAME}.csv"
df.to_csv(out_csv)
print("Saved:", out_csv.resolve())
df.tail()

Out [2]:

Saved: G:\My Drive\Bots DRL\DRL\DRL-MT5-Lab\notebooks\data\ohlc_EURUSD_M15.csv

	open	high	low	close	volume
time
2024-12-31 20:45:00+00:00	1.03568	1.03568	1.03528	1.03531	529
2024-12-31 21:00:00+00:00	1.03528	1.03625	1.03518	1.03619	425
2024-12-31 21:15:00+00:00	1.03618	1.03647	1.03575	1.03590	465
2024-12-31 21:30:00+00:00	1.03590	1.03590	1.03543	1.03560	545
2024-12-31 21:45:00+00:00	1.03560	1.03571	1.03539	1.03542	306

In [ ]:


# Clean shutdown
broker.close_session()

5.9 KiB Raw Blame History

1) Data — Collect & Save OHLC (MT5)

5.9 KiB

Raw Blame History