Create app.py

app.py

@@ -0,0 +1,73 @@
+import torch
+import torch.nn as nn
+import numpy as np
+import gradio as gr
+import joblib
+
+SEQUENCE_LENGTH = 20
+INPUT_SIZE = 1
+OUTPUT_SIZE = 1
+HIDDEN_SIZE = 50
+device = torch.device('cpu')
+
+class SimpleLSTM(nn.Module):
+    def __init__(self, input_size, hidden_size, output_size):
+        super(SimpleLSTM, self).__init__()
+        self.hidden_size = hidden_size
+        self.lstm = nn.LSTM(
+            input_size=input_size,
+            hidden_size=hidden_size,
+            batch_first=True
+        )
+        self.linear = nn.Linear(hidden_size, output_size)
+
+    def forward(self, x):
+        lstm_out, _ = self.lstm(x)
+        last_time_step_output = lstm_out[:, -1, :]
+        prediction = self.linear(last_time_step_output)
+        return prediction
+
+model_path = 'stock_model_weights.pth'
+model = SimpleLSTM(INPUT_SIZE, HIDDEN_SIZE, OUTPUT_SIZE).to(device)
+model.load_state_dict(torch.load(model_path, map_location=device))
+model.eval()
+
+scaler_path = "scaler.joblib"
+scaler = joblib.load(scaler_path)
+
+def predict_stock(input_text):
+    try:
+        numbers = [float(n.strip()) for n in input_text.split(',')]
+        
+        if len(numbers) != SEQUENCE_LENGTH:
+            return f"Error: Please enter exactly {SEQUENCE_LENGTH} numbers."
+        
+        input_data = np.array(numbers).reshape(-1, 1)
+        
+        scaled_input = scaler.transform(input_data)
+        
+        tensor_input = torch.from_numpy(scaled_input).float().reshape(1, SEQUENCE_LENGTH, 1).to(device)
+        
+        with torch.no_grad():
+            scaled_prediction = model(tensor_input)
+            
+        prediction = scaler.inverse_transform(scaled_prediction.cpu().numpy())
+        
+        return f"Predicted Next Day's Price: ${prediction[0][0]:.2f}"
+        
+    except Exception as e:
+        return f"An error occurred: {str(e)}"
+
+demo = gr.Interface(
+    fn=predict_stock,
+    inputs=gr.Textbox(
+        label=f"Input Sequence (Last {SEQUENCE_LENGTH} Days' Close Prices)",
+        placeholder="Enter {SEQUENCE_LENGTH} numbers, separated by commas..."
+    ),
+    outputs=gr.Textbox(label="Prediction"),
+    title="Stock Price Predictor",
+    description="Enter the last 20 closing prices to predict the next day's price.",
+    allow_flagging="never"
+)
+
+demo.launch()