Update utils/model_inference.py

utils/model_inference.py

@@ -1,60 +1,44 @@
 import numpy as np
-import pandas as pd
-from datetime import datetime
+from datetime import datetime, timedelta
 import pytz
-import requests
 
-# Access token for ipinfo.io
-ACCESS_TOKEN = '37b621e95809fa'
-
-# Function to get user timezone based on their IP address
-def get_user_timezone():
-    try:
-        # Get the public IP address of the user
-        response = requests.get(f'https://ipinfo.io?token={ACCESS_TOKEN}')
-        data = response.json()
-        
-        # Extract the timezone information from the response
-        user_timezone = data['timezone']
-        return user_timezone
-    except Exception as e:
-        print(f"Error fetching timezone: {e}")
-        return 'UTC'  # Fallback to UTC if there's an error
-
-# Function to generate forex signals
-def generate_forex_signals(trading_capital, market_risk):
-    # Get the user's timezone based on their IP address
-    user_timezone = get_user_timezone()
-    
+# Function to generate signals for multiple currency pairs
+def generate_forex_signals(trading_capital, market_risk, user_timezone):
     # Ensure the user timezone is valid
     try:
         user_tz = pytz.timezone(user_timezone)
     except pytz.UnknownTimeZoneError:
         raise ValueError("Invalid timezone entered. Please check the format.")
     
-    # Example of how you might process trading capital and risk level:
+    # Define market risk levels and their corresponding risk percentages
     risk_level = {'Low': 0.01, 'Medium': 0.03, 'High': 0.05}
-    
     if market_risk not in risk_level:
-        raise ValueError("Invalid risk level. Choose from Low, Medium, High.")
-    
+        raise ValueError("Invalid risk level. Choose from Low, Medium, or High.")
     risk_percentage = risk_level[market_risk]
-    
-    # Dummy signal generation (Replace with your model inference logic)
-    currency_pair = "EUR/USD"
-    
-    # Get current time in the user's timezone and format it
-    entry_time = datetime.now(user_tz).strftime("%Y-%m-%d %I:%M:%S %p")
-    exit_time = (datetime.now(user_tz) + pd.Timedelta(hours=2)).strftime("%Y-%m-%d %I:%M:%S %p")
-    
-    roi = np.random.uniform(5, 15)  # Random ROI between 5% and 15%
-    signal_strength = np.random.uniform(0.7, 1.0)  # Random strength between 0.7 and 1.0
 
-    # Return the result as a dictionary
+    # Currency pairs to evaluate
+    currency_pairs = ["EUR/USD", "GBP/USD", "USD/JPY"]
+
+    # Generate dummy signals for each currency pair (replace this with your model's predictions)
+    signals = []
+    for pair in currency_pairs:
+        entry_time = datetime.now(user_tz).strftime("%Y-%m-%d %I:%M:%S %p")
+        exit_time = (datetime.now(user_tz) + timedelta(hours=2)).strftime("%Y-%m-%d %I:%M:%S %p")
+        roi = np.random.uniform(5, 20)  # Random ROI between 5% and 20%
+        signal_strength = np.random.uniform(0.7, 1.0)  # Random signal strength
+        signals.append({
+            "currency_pair": pair,
+            "entry_time": entry_time,
+            "exit_time": exit_time,
+            "roi": roi,
+            "signal_strength": signal_strength
+        })
+
+    # Find the signal with the highest ROI
+    best_signal = max(signals, key=lambda x: x["roi"])
+
+    # Return the best signal and all signals
     return {
-        "currency_pair": currency_pair,
-        "entry_time": entry_time,
-        "exit_time": exit_time,
-        "roi": roi,
-        "signal_strength": signal_strength
+        "best_signal": best_signal,
+        "all_signals": signals
     }