Create app.py

app.py

@@ -0,0 +1,66 @@
+import gradio as gr
+import numpy as np
+import pandas as pd
+import tensorflow as tf
+import joblib
+from tensorflow.keras.models import load_model
+
+# Load the trained LSTM model
+model = load_model("lstm_gru_model5.h5")
+
+# Load the MinMaxScaler
+scaler = joblib.load("scaler.pkl")
+
+def preprocess_input(data):
+    """Preprocess input data for LSTM model."""
+    scaled_data = scaler.transform(np.array(data).reshape(-1, 1))
+    return np.array(scaled_data).reshape(1, len(data), 1)  # Reshape for LSTM
+
+def predict_forex(prices):
+    """Predict the next forex price based on the input sequence."""
+    try:
+        input_data = [float(price) for price in prices.split(",")]
+        if len(input_data) < 10:  # Ensure enough input data
+            return "Please provide at least 10 previous forex prices."
+        
+        preprocessed_data = preprocess_input(input_data[-10:])  # Use last 10 prices
+        prediction = model.predict(preprocessed_data)
+        predicted_price = scaler.inverse_transform(prediction)[0][0]  # Convert back to original scale
+        return f"Predicted Next Price: {predicted_price:.5f}"
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+def batch_predict(file):
+    """Batch prediction for CSV files."""
+    try:
+        df = pd.read_csv(file)
+        if "prices" not in df.columns:
+            return "CSV must have a 'prices' column with historical data."
+        
+        df["predictions"] = df["prices"].rolling(window=10).apply(lambda x: predict_forex(",".join(map(str, x))) if len(x) == 10 else None)
+        return df.dropna()
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+# Gradio UI
+demo = gr.Interface(
+    fn=predict_forex,
+    inputs=gr.Textbox(label="Enter last 10 forex prices (comma-separated)"),
+    outputs=gr.Textbox(label="Predicted Next Price"),
+    title="Forex Price Predictor",
+    description="Enter the last 10 forex prices to predict the next price. Upload CSV for batch predictions.",
+    examples=[
+        ["1.2345,1.2350,1.2360,1.2370,1.2380,1.2390,1.2400,1.2410,1.2420,1.2430"]
+    ],
+    allow_flagging="never"
+)
+
+batch_demo = gr.Interface(
+    fn=batch_predict,
+    inputs=gr.File(label="Upload CSV"),
+    outputs=gr.Dataframe(label="Predictions"),
+    title="Batch Prediction",
+    description="Upload a CSV with a 'prices' column for batch predictions."
+)
+
+gr.TabbedInterface([demo, batch_demo], ["Single Prediction", "Batch Prediction"]).launch()