[Bug]: No similarity score returns when calling HTTP/Python API of [Converse with chat assistant]

[valkryhx]

Is there an existing issue for the same bug?

• I have checked the existing issues.

RAGFlow workspace code commit ID

main

RAGFlow image version

main

Other environment information

docker with win11

Actual behavior

I just flow the steps to build a RAG chat assistant ,but it never output the similarity score, a lot of issues about this bug remains. Other parts of chunks are all okay , I searched a lot about this issue ,but I can not find out why . Whether I use a rerank or not , the test of knowledge base is always with similarity scores , but when calling API , no similarity.

Expected behavior

No response

Steps to reproduce

just call the chat assistant with a knowledge base

Additional information

No response

[KevinHuSh]

Could you paste some of your test code and the corresponding response here?

[ams-idk]

Hello @KevinHuSh , I am facing the same issue. Version 0.17.0-slim. Here is the code:

# questions: List[str], session: <ragflow_sdk.modules.session.Session>
chat_config = {
name = name, dataset_ids = dataset_ids, llm=Chat.LLM(rag=True, res_dict={"model_name":chat, temperature": 0.1, "top_p": 0.3, presence_penalty":0.2, "frequency_penalty":0.7, "max_tokens": 512}, prompt=Chat.Prompt(rag=True,res_dict={similarity_threshold":0.2,"keywords_similarity_weight": 0.7,"top_n":8,"variables":  [{"key": "knowledge", "optional": True}],"rerank_model":"","top_k": 1024,"empty_response": None,"opener": "Hello, how can I help you today?", "show_quote": True, "prompt": "You are an intelligent assistant. Please summarize the content of the knowledge base to answer the question. Please list the data in the knowledge base and answer in detail. When all knowledge base content is irrelevant to the question, your answer must include the sentence 'The answer you are looking for is not found in the knowledge base!' Answers need to consider chat history. Here is the knowledge base:{knowledge}The above is the knowledge base."
}
assistant = rag_object.create_chat(**chat_config)
session = assistant.create_session()
respuesta = {}
for question in questions:
    for ans in session.ask(question, stream=False):
        respuesta['Respuesta'] = ans.content
        respuesta['Chunks'] = ans.reference
        respuesta['Pregunta'] = question
        respuestas_preguntas.append(respuesta)

The response is below. You can see how there is no similiarity info included.