Spaces:
Running
Running
| """Answer questions about my resume.""" | |
| # %% IMPORTS | |
| import logging | |
| import gradio as gr | |
| import lib | |
| # %% LOGGING | |
| logging.basicConfig( | |
| level=logging.INFO, | |
| format="[%(asctime)s][%(levelname)s] %(message)s", | |
| ) | |
| # %% CONFIGS | |
| # %% - Frontend | |
| THEME = "soft" | |
| TITLE = "Fmind AI Assistant" | |
| EXAMPLES = [ | |
| "Who is Médéric Hurier (Fmind)?", | |
| "Is Fmind open to new opportunities?", | |
| "Can you share details about Médéric PhD?", | |
| "Elaborate on Médéric current work position", | |
| "Describe his proficiency with Python programming", | |
| "What is the answer to life, the universe, and everything?", | |
| ] | |
| # %% - Backend | |
| MODEL = lib.get_language_model() | |
| CLIENT = lib.get_database_client(path=lib.DATABASE_PATH) | |
| ENCODING = lib.get_encoding_function() | |
| EMBEDDING = lib.get_embedding_function() | |
| COLLECTION = CLIENT.get_collection( | |
| name=lib.DATABASE_COLLECTION, | |
| embedding_function=EMBEDDING, | |
| ) | |
| # %% - Answer | |
| PROMPT_CONTEXT = """ | |
| You are Fmind AI Assistant, specialized in providing information regarding Médéric Hurier's (known as Fmind) professional background. | |
| Médéric is a Lead MLOps engineer based in Luxembourg. He is currently working at Decathlon. His calendar is booked until the conclusion of 2024. | |
| Your responses should be succinct and maintain a professional tone. If inquiries deviate from Médéric's professional sphere, courteously decline to engage. | |
| You may find more information about Médéric below (markdown format): | |
| """ | |
| PROMPT_MAX_TOKENS = lib.MODEL_INPUT_LIMIT | |
| QUERY_MAX_DISTANCE = 0.4 | |
| QUERY_N_RESULTS = 20 | |
| # %% FUNCTIONS | |
| def answer(message: str, history: list[str]) -> str: | |
| """Answer questions about my resume.""" | |
| # counters | |
| n_tokens = 0 | |
| # messages | |
| messages = [] | |
| # - context | |
| n_tokens += len(ENCODING(PROMPT_CONTEXT)) | |
| messages += [{"role": "system", "content": PROMPT_CONTEXT}] | |
| # - history | |
| for user_content, assistant_content in history: | |
| n_tokens += len(ENCODING(user_content)) | |
| n_tokens += len(ENCODING(assistant_content)) | |
| messages += [{"role": "user", "content": user_content}] | |
| messages += [{"role": "assistant", "content": assistant_content}] | |
| # - message | |
| n_tokens += len(ENCODING(message)) | |
| messages += [{"role": "user", "content": message}] | |
| # database | |
| results = COLLECTION.query(query_texts=message, n_results=QUERY_N_RESULTS) | |
| logging.info("Results: %s", results) | |
| distances = results["distances"][0] | |
| documents = results["documents"][0] | |
| for distance, document in zip(distances, documents): | |
| # - distance | |
| logging.debug("Doc distance: %f", distance) | |
| if distance > QUERY_MAX_DISTANCE: | |
| break | |
| # - document | |
| n_document_tokens = len(ENCODING(document)) | |
| logging.debug("Doc tokens: %f", n_document_tokens) | |
| if (n_tokens + n_document_tokens) >= PROMPT_MAX_TOKENS: | |
| break | |
| n_tokens += n_document_tokens | |
| messages[0]["content"] += document | |
| # response | |
| logging.info("Tokens: %d", n_tokens) | |
| logging.info("Messages: %s", messages) | |
| api_response = MODEL(messages=messages) | |
| logging.info("Response: %s", api_response.to_dict_recursive()) | |
| # content | |
| content = api_response["choices"][0]["message"]["content"] | |
| # return | |
| return content | |
| # %% INTERFACES | |
| interface = gr.ChatInterface( | |
| fn=answer, | |
| theme=THEME, | |
| title=TITLE, | |
| examples=EXAMPLES, | |
| clear_btn=None, | |
| retry_btn=None, | |
| undo_btn=None, | |
| ) | |
| if __name__ == "__main__": | |
| interface.queue(concurrency_count=20).launch() | |