Create app.py

app.py

@@ -0,0 +1,59 @@
+# app.py
+# বাংলা মন্তব্য সহ Hugging Face Space deploy-ready কোড
+
+import gradio as gr
+import torch
+import torch.nn as nn
+
+# -----------------------------
+# Hybrid Model Design
+# -----------------------------
+class HybridModel(nn.Module):
+    def __init__(self, input_size=1, hidden_size=64, num_layers=2):
+        super(HybridModel, self).__init__()
+        # LSTM sequence শেখার জন্য
+        self.lstm = nn.LSTM(input_size, hidden_size, num_layers, batch_first=True)
+        # Transformer context শেখার জন্য
+        encoder_layer = nn.TransformerEncoderLayer(d_model=hidden_size, nhead=4)
+        self.transformer = nn.TransformerEncoder(encoder_layer, num_layers=2)
+        # Output layer
+        self.fc = nn.Linear(hidden_size, 1)
+
+    def forward(self, x):
+        lstm_out, _ = self.lstm(x)
+        trans_out = self.transformer(lstm_out)
+        out = self.fc(trans_out[:, -1, :])
+        return out
+
+# Dummy model (শুধু demo এর জন্য)
+model = HybridModel()
+
+# -----------------------------
+# Prediction Function
+# -----------------------------
+def predict(input_value):
+    try:
+        val = float(input_value)
+        x = torch.tensor([[[val]]], dtype=torch.float32)
+        with torch.no_grad():
+            pred = model(x).item()
+        return f"🔮 Prediction: {pred:.2f}"
+    except:
+        return "❌ Invalid input! Please enter a number like 2.1"
+
+# -----------------------------
+# Gradio UI
+# -----------------------------
+with gr.Blocks(theme=gr.themes.Monochrome()) as demo:
+    gr.Markdown("## 🟣 Real-Time Hybrid AI Prediction System\nবাংলায় সহজ UI, Dark Mode")
+
+    with gr.Row():
+        input_box = gr.Textbox(label="Live Data Input (e.g. 2.1)", placeholder="Type number ➕ ⬜ ➖ ▫️ ➕ ⬜ ➖")
+    output_box = gr.Textbox(label="Prediction Result", interactive=False)
+
+    input_box.change(fn=predict, inputs=input_box, outputs=output_box)
+
+    # CSV Upload Option
+    csv_file = gr.File(label="Upload CSV for Initial Learning", file_types=[".csv"])
+
+demo.launch()