Deploy Ericsson LLM chatbot with RAG

app.py

@@ -0,0 +1,63 @@
+
+from langchain.document_loaders import UnstructuredPDFLoader, CSVLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.chains import RetrievalQA
+from langchain.llms import HuggingFaceHub
+import gradio as gr
+from PIL import Image
+import pytesseract
+
+repo_id = "meta-llama/Llama-3.2-11B-Vision-Instruct"
+llm = HuggingFaceHub(repo_id=repo_id, model_kwargs={"temperature": 0.7})
+
+def load_documents(files):
+    docs = []
+    for file_path in files:
+        if file_path.endswith(".pdf") or file_path.endswith(".pptx"):
+            loader = UnstructuredPDFLoader(file_path)
+            docs.extend(loader.load())
+        elif file_path.endswith(".csv"):
+            loader = CSVLoader(file_path)
+            docs.extend(loader.load())
+        elif file_path.endswith((".jpg", ".png")):
+            img = Image.open(file_path)
+            text = pytesseract.image_to_string(img, lang="kor+eng")
+            docs.append(text)
+    splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
+    texts = splitter.split_documents(docs)
+    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/multi-qa-mpnet-base-dot-v1")
+    vectorstore = FAISS.from_documents(texts, embeddings)
+    return vectorstore
+
+def create_rag_chain(vectorstore):
+    qa_chain = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=vectorstore.as_retriever(search_kwargs={"k": 3})
+    )
+    return qa_chain
+
+def chatbot(query, files):
+    if not files:
+        return "파일 업로드 필요 (PDF/CSV/이미지/PPTX)."
+    try:
+        vectorstore = load_documents(files)
+        qa_chain = create_rag_chain(vectorstore)
+        response = qa_chain.run(query)
+        return response
+    except Exception as e:
+        return f"오류: {str(e)}."
+
+with gr.Blocks(title="Ericsson 장비 분석 챗봇") as demo:
+    gr.Markdown("# 🚀 3G/LTE/5G 장비 불량/불요파 분석")
+    gr.Markdown("PDF/CSV/이미지 업로드 후 질문: e.g., 'Spurious Emission 통계?'")
+    query = gr.Textbox(label="질문 (한국어/영어)", placeholder="장애 원인 분석해줘")
+    files = gr.File(label="파일 업로드", file_count="multiple")
+    output = gr.Textbox(label="응답", lines=10)
+    btn = gr.Button("분석 시작!")
+    btn.click(chatbot, inputs=[query, files], outputs=output)
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,14 @@
+
+transformers
+torch
+langchain
+gradio
+unstructured[all-docs]
+faiss-cpu
+peft
+datasets
+pandas
+pytesseract
+pillow
+sentence-transformers
+huggingface_hub