coding

app.py

@@ -2,6 +2,8 @@ import streamlit as st
 from langchain_community.llms import HuggingFaceHub
 from langchain_community.embeddings import SentenceTransformerEmbeddings
 from langchain_community.vectorstores import FAISS
+from datasets import load_dataset
+import random
 
 # 1. 准备知识库数据 (示例)
 knowledge_base = [
@@ -12,10 +14,19 @@ knowledge_base = [
     "Gemma 支持多种语言。"
 ]
 
+try:
+    dataset = load_dataset("rorubyy/attack_on_titan_wiki_chinese")
+    answer_list = [example["Answer"] for example in dataset["train"]]
+
+except Exception as e:
+    st.error(f"读取数据集失败：{e}")
+    st.stop()
+
+
 # 2. 构建向量数据库 (如果需要，仅构建一次)
 try:
     embeddings = SentenceTransformerEmbeddings(model_name="all-mpnet-base-v2")
-    db = FAISS.from_texts(knowledge_base, embeddings)
+    db = FAISS.from_texts(answer_list, embeddings)
 except Exception as e:
     st.error(f"向量数据库构建失败：{e}")
     st.stop()
@@ -56,6 +67,20 @@ temperature = st.number_input("temperature", value=1.0)
 max_length = st.number_input("max_length", value=1024)
 question = st.text_area("请输入问题", "Gemma 有哪些特点？")
 
+if st.button("随机"):
+    dataset_size = len(dataset["train"])
+    random_index = random.randint(0, dataset_size - 1)
+    # 读取随机问题
+    random_question = dataset["train"][random_index]["Question"]
+    origin_answer = dataset["train"][random_index]["Answer"]
+    st.write("随机问题：")
+    st.write(random_question)
+    st.write("原始答案：")
+    st.write(origin_answer)
+    answer = answer_question(gemma, float(temperature), int(max_length), random_question)
+    st.write("生成答案：")
+    st.write(answer)
+
 if st.button("提交"):
     if not question:
         st.warning("请输入问题！")

requirements.txt

@@ -5,3 +5,4 @@ langchain-community
 langchain-huggingface
 sentence_transformers
 faiss-cpu
+datasets