Created app.py with CodeLlama-3b model

app.py

@@ -0,0 +1,69 @@
+import streamlit as st
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+import faiss
+import numpy as np
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+
+# 🎯 Load CodeLlama 3B Model
+model_name = "codellama/CodeLlama-3b"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
+
+# 🔍 Function to Generate Responses
+def generate_response(prompt):
+    inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+    output = model.generate(**inputs, max_length=300)
+    return tokenizer.decode(output[0], skip_special_tokens=True)
+
+# 🔍 FAISS Index Creation
+def create_faiss_index(texts):
+    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
+    embedding_vectors = embeddings.embed_documents(texts)
+    faiss_index = faiss.IndexFlatL2(len(embedding_vectors[0]))
+    faiss_index.add(np.array(embedding_vectors))
+    return faiss_index
+
+# 📂 Function to Parse Logs
+def parse_test_log(file):
+    try:
+        log_content = file.read().decode("utf-8")
+        return log_content.splitlines()  # Split logs into lines
+    except Exception as e:
+        st.error(f"Error reading file: {e}")
+        return []
+
+# 🌟 Streamlit UI
+st.title("🔍 Test Log Failure Analysis with CodeLlama 3B")
+
+# 📂 File Upload
+uploaded_file = st.file_uploader("Upload test log (txt, json, xml, html)", type=["txt", "json", "xml", "html"])
+
+if uploaded_file is not None:
+    st.info("Processing the test log file...")
+    test_log_lines = parse_test_log(uploaded_file)
+
+    # 🔍 Create FAISS Index
+    st.info("Indexing log data...")
+    faiss_index = create_faiss_index(test_log_lines)
+
+    # ❓ User Query
+    question = st.text_input("Ask a question about the test failures:")
+
+    if question:
+        # 🔍 Retrieve Similar Logs from FAISS
+        query_embedding = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2").embed_query(question)
+        _, I = faiss_index.search(np.array([query_embedding]), k=5)  # Retrieve top 5 results
+
+        # 📝 Combine the most relevant logs
+        context = "\n".join([test_log_lines[i] for i in I[0]])
+
+        # 🤖 Generate Answer with CodeLlama
+        prompt = f"Given the following test logs:\n{context}\n\nAnswer: {question}"
+        answer = generate_response(prompt)
+        st.subheader("Analysis Result:")
+        st.write(answer)
+
+else:
+    st.info("Upload a test log file to begin analysis.")