ExternalVendorCode/lollms-webui
1
0
Fork 0
mirror of https://github.com/ParisNeo/lollms-webui.git synced 2025-02-11 13:05:22 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

moved to threading instead of threadpools

Browse Source
This commit is contained in:
ParisNeo 2023-04-24 22:02:50 +02:00
parent 917600f412
commit 79426ce9d0
2 changed files with 24 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
app.py
View File

@ -20,7 +20,7 @@ import argparse
import json import json
import re import re
import traceback import traceback
from concurrent.futures import ThreadPoolExecutor import threading
import sys import sys
from pyGpt4All.db import DiscussionsDB, Discussion from pyGpt4All.db import DiscussionsDB, Discussion
from flask import ( from flask import (
@ -263,8 +263,10 @@ class Gpt4AllWebUI(GPT4AllAPI):
self.discussion_messages = self.prepare_query(message_id) self.discussion_messages = self.prepare_query(message_id)
self.prepare_reception() self.prepare_reception()
self.generating = True self.generating = True
app.config['executor'] = ThreadPoolExecutor(max_workers=1) # app.config['executor'] = ThreadPoolExecutor(max_workers=1)
app.config['executor'].submit(self.generate_message) # app.config['executor'].submit(self.generate_message)
tpe = threading.Thread(target=self.generate_message)
tpe.start()
while self.generating: while self.generating:
try: try:
while not self.text_queue.empty(): while not self.text_queue.empty():
@ -279,7 +281,8 @@ class Gpt4AllWebUI(GPT4AllAPI):
time.sleep(0.1) time.sleep(0.1)
if self.cancel_gen: if self.cancel_gen:
self.generating = False self.generating = False
app.config['executor'].shutdown(True) tpe = None
gc.collect()
print("## Done ##") print("## Done ##")
self.current_discussion.update_message(response_id, self.bot_says) self.current_discussion.update_message(response_id, self.bot_says)
self.full_message_list.append(self.bot_says) self.full_message_list.append(self.bot_says)
@ -583,8 +586,8 @@ if __name__ == "__main__":
personality = load_config(f"personalities/{config['personality_language']}/{config['personality_category']}/{config['personality']}.yaml") personality = load_config(f"personalities/{config['personality_language']}/{config['personality_category']}/{config['personality']}.yaml")
executor = ThreadPoolExecutor(max_workers=1) # executor = ThreadPoolExecutor(max_workers=1)
app.config['executor'] = executor # app.config['executor'] = executor
bot = Gpt4AllWebUI(app, config, personality, config_file_path) bot = Gpt4AllWebUI(app, config, personality, config_file_path)

3
backends/llama_cpp/__init__.py
View File

@ -50,6 +50,7 @@ class LLAMACPP(GPTBackend):
new_text_callback (Callable[[str], None], optional): A callback function that is called everytime a new text element is generated. Defaults to None. new_text_callback (Callable[[str], None], optional): A callback function that is called everytime a new text element is generated. Defaults to None.
verbose (bool, optional): If true, the code will spit many informations about the generation process. Defaults to False. verbose (bool, optional): If true, the code will spit many informations about the generation process. Defaults to False.
""" """
try:
self.model.generate( self.model.generate(
prompt, prompt,
new_text_callback=new_text_callback, new_text_callback=new_text_callback,
@ -62,3 +63,5 @@ class LLAMACPP(GPTBackend):
n_threads=self.config['n_threads'], n_threads=self.config['n_threads'],
verbose=verbose verbose=verbose
) )
except Exception as ex:
print(ex)