Create app.py

app.py

@@ -0,0 +1,40 @@
+import streamlit as st
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+from peft import PeftModel
+import torch
+
+@st.cache_resource
+def load_model():
+    model_id = "google/flan-t5-large"
+    adapter_path = "./Flan-T5-Typosquat-detect"  # Adjust to your saved adapter path
+
+    # Load the tokenizer and model
+    tokenizer = AutoTokenizer.from_pretrained(model_id)
+    model = AutoModelForSeq2SeqLM.from_pretrained(model_id)
+    model = PeftModel.from_pretrained(model, adapter_path)
+    model.eval()
+    
+    return model, tokenizer
+
+model, tokenizer = load_model()
+
+st.title("FLAN-T5 Typosquatting Detection")
+st.write("Enter a potential typosquatted domain and a target domain to check if one is a variant of the other.")
+
+potential_typosquat = st.text_input("Potential Typosquatted Domain", value="lonlonsoft.com")
+target_domain = st.text_input("Target Domain", value="stiltsoft.net")
+
+if st.button("Check Typosquatting"):
+    if potential_typosquat and target_domain:
+        
+        prompt = f"Is the first domain a typosquat of the second: {potential_typosquat} {target_domain}"
+
+        input_ids = tokenizer(prompt, return_tensors="pt").input_ids
+        outputs = model.generate(input_ids, max_new_tokens=20)
+        
+        prediction = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        
+        st.write("**Prediction:**")
+        st.write(prediction)
+    else:
+        st.warning("Please enter both domains to perform the check.")