correcao de dependencia

app.py

@@ -6,7 +6,7 @@ from transformers import pipeline
 from langchain.llms.base import LLM
 from llama_index import SimpleDirectoryReader, GPTVectorStoreIndex, PromptHelper, LLMPredictor, ServiceContext
 from llama_index.langchain_helpers.text_splitter import TokenTextSplitter
-from llama_index.node_parser import SimpleNodeParser
+from llama_index.node_parser import SentenceSplitter
 from langchain.embeddings.huggingface import HuggingFaceEmbeddings
 
 from llama_index.embeddings import LangchainEmbedding
@@ -17,8 +17,9 @@ index_files = list(Path(INPUT_FOLDER).glob("*"))
 
 max_input_size = 2048
 num_output = 256
-max_chunk_overlap = 0.5
-prompt_helper = PromptHelper(max_input_size, num_output, max_chunk_overlap)
+max_chunk_overlap = 20
+max_prompt_chunk_overlap = 0.5
+prompt_helper = PromptHelper(max_input_size, num_output, max_prompt_chunk_overlap)
 
 pipe = pipeline("text-generation", model="databricks/dolly-v2-3b", trust_remote_code=True, torch_dtype=torch.bfloat16, device_map="auto")
 embed_model = LangchainEmbedding(HuggingFaceEmbeddings())
@@ -41,8 +42,8 @@ class CustomLLM(LLM):
 # define our LLM
 llm_predictor = LLMPredictor(llm=CustomLLM())
 
-node_parser = SimpleNodeParser(text_splitter=TokenTextSplitter(chunk_size=512, chunk_overlap=max_chunk_overlap))
-prompt_helper = PromptHelper(max_input_size, num_output, max_chunk_overlap)
+node_parser = SentenceSplitter(chunk_size=512, chunk_overlap=max_chunk_overlap)
+prompt_helper = PromptHelper(max_input_size, num_output, max_prompt_chunk_overlap)
 service_context = ServiceContext.from_defaults(llm_predictor=llm_predictor, embed_model=embed_model, prompt_helper=prompt_helper, node_parser=node_parser, chunk_size_limit=512)
 # Load your data
 documents = SimpleDirectoryReader(input_files=index_files).load_data()
@@ -55,4 +56,4 @@ def greet(query):
     return query_engine.query(query)
 
 iface = gr.Interface(fn=greet, inputs="text", outputs="text")
-iface.launch()
+iface.launch(share=True)