Spaces:
Running
Running
File size: 5,263 Bytes
78efe79 440418c f3985af dc80b35 22dee1c fe505c6 104f578 76119f7 407a575 9c071a8 c4be42e 9c071a8 c4be42e 9c071a8 c4be42e 9c071a8 104f578 76119f7 104f578 9c071a8 76119f7 32c38ef f3985af 440418c 1831164 440418c 22dee1c 440418c 22dee1c 08baccf 104f578 dc80b35 40d0e92 74ccf1c 12bb502 78efe79 08baccf dc80b35 08baccf 78efe79 40d0e92 dc80b35 78efe79 5f46973 aab7a25 5f46973 13feae4 5f46973 9c071a8 b4a00e6 5f46973 aab7a25 5f46973 aab7a25 5f46973 9c071a8 104f578 9c071a8 76119f7 104f578 76119f7 104f578 9c071a8 104f578 9c071a8 b4a00e6 aab7a25 34428f1 dc80b35 76119f7 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 |
import discord
import logging
import os
from huggingface_hub import InferenceClient
import asyncio
import subprocess
from datasets import load_dataset
import pandas as pd
from fuzzywuzzy import process
# ํ์ฌ ์์
๋๋ ํ ๋ฆฌ ์ถ๋ ฅ
print("Current Working Directory:", os.getcwd())
# ๋ฐ์ดํฐ์
ํ์ผ ์ด๋ฆ
data_files = ['train_0.csv', 'train_1.csv', 'train_2.csv', 'train_3.csv', 'train_4.csv', 'train_5.csv']
# ํ์ฌ ์์
๋๋ ํ ๋ฆฌ์ ๋ชจ๋ ํ์ผ์ด ์๋์ง ํ์ธ
missing_files = [file for file in data_files if not os.path.exists(file)]
if missing_files:
print(f"Missing files: {missing_files}")
# ํ์ํ ๊ฒฝ์ฐ ์์
๋๋ ํ ๋ฆฌ ๋ณ๊ฒฝ
os.chdir('/home/user/app')
print("Changed directory to:", os.getcwd())
else:
print("All files are present in the current directory.")
# ๋ฐ์ดํฐ์
๋ก๋ ๋ฐ ์ต์ ํ
def load_optimized_dataset(data_files):
data_frames = [pd.read_csv(file) for file in data_files]
full_data = pd.concat(data_frames, ignore_index=True)
# ์ฌ๊ฑด๋ช
์ ํค๋ก ํ๊ณ ์ฌ๊ฑด๋ฒํธ์ ์ ๋ฌธ์ ์ ์ฅํ๋ ๋์
๋๋ฆฌ ์์ฑ
name_to_number = full_data.groupby('์ฌ๊ฑด๋ช
')['์ฌ๊ฑด๋ฒํธ'].apply(list).to_dict()
number_to_fulltext = full_data.set_index('์ฌ๊ฑด๋ฒํธ')['์ ๋ฌธ'].to_dict()
return name_to_number, number_to_fulltext
name_to_number, number_to_fulltext = load_optimized_dataset(data_files)
print("Dataset loaded successfully.")
# ์ฌ๊ฑด๋ช
๋ฆฌ์คํธ ์์ฑ
all_case_names = list(name_to_number.keys())
# ๋ก๊น
์ค์
logging.basicConfig(level=logging.DEBUG, format='%(asctime)s:%(levelname)s:%(name)s: %(message)s', handlers=[logging.StreamHandler()])
# ์ธํ
ํธ ์ค์
intents = discord.Intents.default()
intents.message_content = True
intents.messages = True
intents.guilds = True
intents.guild_messages = True
# ์ถ๋ก API ํด๋ผ์ด์ธํธ ์ค์ (์์๋ก ํฌํจ, ์ค์ ๋ก ์ฌ์ฉ๋์ง ์์)
hf_client = InferenceClient("CohereForAI/c4ai-command-r-plus", token=os.getenv("HF_TOKEN"))
# ํน์ ์ฑ๋ ID
SPECIFIC_CHANNEL_ID = int(os.getenv("DISCORD_CHANNEL_ID"))
# ๋ํ ํ์คํ ๋ฆฌ๋ฅผ ์ ์ฅํ ์ ์ญ ๋ณ์
conversation_history = []
class MyClient(discord.Client):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.is_processing = False
async def on_ready(self):
logging.info(f'{self.user}๋ก ๋ก๊ทธ์ธ๋์์ต๋๋ค!')
subprocess.Popen(["python", "web.py"])
logging.info("Web.py server has been started.")
async def on_message(self, message):
if message.author == self.user:
return
if not self.is_message_in_specific_channel(message):
return
if self.is_processing:
logging.debug("Currently processing another message, skipping this one.")
return
self.is_processing = True
try:
response = await generate_response(message)
if response and response.strip():
await message.channel.send(response)
else:
await message.channel.send("์ฃ์กํฉ๋๋ค, ์ ๊ณตํ ์ ์๋ ์ ๋ณด๊ฐ ์์ต๋๋ค.")
finally:
self.is_processing = False
logging.debug("Message processing completed, ready for the next one.")
def is_message_in_specific_channel(self, message):
channel_condition = message.channel.id == SPECIFIC_CHANNEL_ID
thread_condition = isinstance(message.channel, discord.Thread) and message.channel.parent_id == SPECIFIC_CHANNEL_ID
return channel_condition or thread_condition
async def generate_response(message):
global conversation_history
user_input = message.content.strip()
user_mention = message.author.mention
# ์ ์ฌํ ์ฌ๊ฑด๋ช
์ฐพ๊ธฐ
matched_case_names = process.extractBests(user_input, all_case_names, limit=3, score_cutoff=80)
if matched_case_names:
case_numbers = []
for case_name, score in matched_case_names:
case_numbers.extend(name_to_number[case_name])
case_numbers = list(set(case_numbers)) # ์ค๋ณต ์ ๊ฑฐ
case_numbers_str = "\n".join(case_numbers)
system_message = f"{user_mention}, '{user_input}'์ ์ ์ฌํ ์ฌ๊ฑด๋ช
์ ์ฌ๊ฑด๋ฒํธ๋ ๋ค์๊ณผ ๊ฐ์ต๋๋ค:\n{case_numbers_str}"
elif user_input in number_to_fulltext:
full_text = number_to_fulltext[user_input]
system_message = f"{user_mention}, ์ฌ๊ฑด๋ฒํธ '{user_input}'์ ์ ๋ฌธ์ ๋ค์๊ณผ ๊ฐ์ต๋๋ค:\n\n{full_text}"
else:
system_message = f"{user_mention}, ๊ด๋ จ ๋ฒ๋ฅ ์ ๋ณด๋ฅผ ์ฐพ์ ์ ์์ต๋๋ค."
# ๋ฉ์์ง ๊ธธ์ด ์ ํ ์ฒ๋ฆฌ
max_length = 2000
if len(system_message) > max_length:
response_parts = []
for i in range(0, len(system_message), max_length):
part_response = system_message[i:i + max_length]
await message.channel.send(part_response)
response_parts.append(part_response)
return response_parts[0] if response_parts else "์ฃ์กํฉ๋๋ค, ์ ๊ณตํ ์ ์๋ ์ ๋ณด๊ฐ ์์ต๋๋ค."
return system_message
if __name__ == "__main__":
discord_client = MyClient(intents=intents)
discord_client.run(os.getenv('DISCORD_TOKEN')) |