Create app.py

app.py

@@ -0,0 +1,31 @@
+from langchain.llms import HuggingFaceHub
+from langchain.embeddings import SentenceTransformerEmbeddings
+from langchain.vectorstores import FAISS
+
+# 1. 初始化 Gemma 模型
+llm = HuggingFaceHub(repo_id="google/gemma-7b-it", model_kwargs={"temperature": 0.5, "max_length": 512})
+
+# 2. 准备知识库数据
+knowledge_base = [
+    "Gemma 是 Google 开发的大型语言模型。",
+    "Gemma 具有强大的自然语言处理能力。",
+    "Gemma 可以用于问答、对话、文本生成等任务。"
+]
+
+# 3. 构建向量数据库
+embeddings = SentenceTransformerEmbeddings(model_name="all-mpnet-base-v2")
+db = FAISS.from_texts(knowledge_base, embeddings)
+
+# 4. 问答函数
+def answer_question(question):
+    question_embedding = embeddings.embed_query(question)
+    docs_and_scores = db.similarity_search_with_score(question_embedding)
+    context = "\n".join([doc.page_content for doc, _ in docs_and_scores])
+    prompt = f"请根据以下知识库回答问题：\n{context}\n问题：{question}"
+    answer = llm(prompt)
+    return answer
+
+# 5. 测试
+question = "Gemma 有哪些特点？"
+answer = answer_question(question)
+print(answer)