added streaming output to Gradio

app.py

@@ -6,7 +6,7 @@ from botocore.client import Config
 from langchain.document_loaders import WebBaseLoader
 
 from langchain.text_splitter import RecursiveCharacterTextSplitter
-text_splitter = RecursiveCharacterTextSplitter(chunk_size=350, chunk_overlap=10)
+text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
 
 from langchain.llms import HuggingFaceHub
 model_id = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={"temperature":0.1, "max_new_tokens":1024})
@@ -17,25 +17,16 @@ embeddings = HuggingFaceHubEmbeddings()
 from langchain.vectorstores import Chroma
 
 from langchain.chains import RetrievalQA
-from langchain.chains import RetrievalQAWithSourcesChain
-
-from langchain.prompts import ChatPromptTemplate
-
-#web_links = ["https://www.databricks.com/","https://help.databricks.com","https://docs.databricks.com","https://kb.databricks.com/","http://docs.databricks.com/getting-started/index.html","http://docs.databricks.com/introduction/index.html","http://docs.databricks.com/getting-started/tutorials/index.html","http://docs.databricks.com/machine-learning/index.html","http://docs.databricks.com/sql/index.html"]
-#loader = WebBaseLoader(web_links)
-#documents = loader.load()
 
 s3 = boto3.client('s3', config=Config(signature_version=UNSIGNED))
 s3.download_file('rad-rag-demos', 'vectorstores/chroma.sqlite3', './chroma_db/chroma.sqlite3')
      
 db = Chroma(persist_directory="./chroma_db", embedding_function=embeddings)
 db.get()
-#texts = text_splitter.split_documents(documents)
-#db = Chroma.from_documents(texts, embedding_function=embeddings)
 retriever = db.as_retriever()
 
 global qa 
-qa = RetrievalQAWithSourcesChain.from_chain_type(llm=model_id, chain_type="stuff", retriever=retriever)
+qa = RetrievalQA.from_chain_type(llm=model_id, chain_type="stuff", retriever=retriever)
 
 
 def add_text(history, text):
@@ -44,14 +35,16 @@ def add_text(history, text):
 
 def bot(history):
     response = infer(history[-1][0])
-    history[-1][1] = response['result']
-    return history
+    history[-1][1] = ""
+    for character in response['result']:
+        history[-1][1] += character
+        time.sleep(0.05)
+        yield history
 
 def infer(question):
     
-    #query = question
-    #result = qa({"query": query})
-    result = qa({"question": question})
+    query = question
+    result = qa({"query": query})
     return result
 
 css="""