Spaces:
Running
Running
File size: 3,556 Bytes
beb32ed ed67987 34ba0fa 114ee0c ed67987 5c68cc7 ed67987 5c68cc7 ed67987 beb32ed ed67987 beb32ed 5c68cc7 ed67987 114ee0c ed67987 beb32ed ed67987 beb32ed d37f75a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 |
"""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()
|