Update app.py

app.py

@@ -8,15 +8,24 @@ from transformers import AutoTokenizer
 
 bshtml_dir_loader = DirectoryLoader('./data/', loader_cls=BSHTMLLoader)
 data = bshtml_dir_loader.load()
+print("loading documents")
 
 bloomz_tokenizer = AutoTokenizer.from_pretrained("bigscience/bloomz-1b7")
 
+print("add tokenizer")
+
 text_splitter = CharacterTextSplitter.from_huggingface_tokenizer(bloomz_tokenizer,
                                                    chunk_size=100,
                                                    chunk_overlap=0,
                                                    separator="\n")
+
+
+print("Add text spliters")
+
 documents = text_splitter.split_documents(data)
 
+print("Getting HF embeddings")
+
 embeddings = HuggingFaceEmbeddings()
 
 llm = HuggingFacePipeline.from_model_id(
@@ -24,12 +33,24 @@ llm = HuggingFacePipeline.from_model_id(
         task="text-generation",
         model_kwargs={"temperature" : 0, "max_length" : 500})
 
+print("Adding LLM hugginFacePipeline with bigscience bloomz")
+
 vectordb = Chroma.from_documents(documents=documents, embedding=embeddings)
+
+print("Getting vectors")
+
 doc_retriever = vectordb.as_retriever()
+
+print("Creating Retreiver")
+
+
 shakespeare_qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=doc_retriever)
 
+print("Add shakespeare qa")
+
 def query(query):
     shakespeare_qa.run(query)
+    
 
 iface = gr.Interface(fn=query, inputs="text", outputs="text")
 iface.launch()