Create app.py

app.py

@@ -0,0 +1,32 @@
+import gradio as gr
+from tensorflow.keras.models import load_model
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+import pickle
+from huggingface_hub import hf_hub_download
+
+# Download files from model repo
+model_path = hf_hub_download("lokas/spam-emails-classifier", "model.h5")
+tokenizer_path = hf_hub_download("lokas/spam-emails-classifier", "tokenizer.pkl")
+
+# Load model and tokenizer
+model = load_model(model_path)
+with open(tokenizer_path, "rb") as f:
+    tokenizer = pickle.load(f)
+
+SEQUENCE_LENGTH = 50  # Must match training
+
+def predict_spam(text):
+    seq = tokenizer.texts_to_sequences([text])
+    padded = pad_sequences(seq, maxlen=SEQUENCE_LENGTH)
+    pred = model.predict(padded)[0][0]
+    return "🚫 Spam" if pred > 0.5 else "✅ Not Spam"
+
+interface = gr.Interface(
+    fn=predict_spam,
+    inputs=gr.Textbox(lines=3, placeholder="Paste an email message..."),
+    outputs="text",
+    title="Spam Email Detector",
+    description="A BiLSTM-based spam classifier trained on the Enron dataset with GloVe embeddings."
+)
+
+interface.launch()