rixty
/
KoboldAI-Client
mirror of https://github.com/KoboldAI/KoboldAI-Client.git
1
1
Fork 0
Code Issues Projects Releases Wiki Activity

Fix for web based model loading

This commit is contained in:
ebolam 2022-06-06 12:49:40 -04:00
parent 1e139594a9
commit c984f4412d
2 changed files with 17 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
aiserver.py
View File

@ -53,6 +53,7 @@ from utils import debounce
import utils import utils
import structures import structures
import torch import torch
from transformers import StoppingCriteria, GPT2TokenizerFast, GPT2LMHeadModel, GPTNeoForCausalLM, GPTNeoModel, AutoModelForCausalLM, AutoTokenizer
if lupa.LUA_VERSION[:2] != (5, 4): if lupa.LUA_VERSION[:2] != (5, 4):
@ -321,6 +322,12 @@ class vars:
utils.vars = vars utils.vars = vars
class Send_to_socketio(object):
def write(self, bar):
print(bar, end="")
time.sleep(0.01)
emit('from_server', {'cmd': 'model_load_status', 'data': bar}, broadcast=True)
# Set logging level to reduce chatter from Flask # Set logging level to reduce chatter from Flask
import logging import logging
log = logging.getLogger('werkzeug') log = logging.getLogger('werkzeug')
@ -986,6 +993,10 @@ def load_model(use_gpu=True, key='', gpu_layers=None, initial_load=False):
vars.noai = False vars.noai = False
if not initial_load: if not initial_load:
set_aibusy(True) set_aibusy(True)
if vars.model != 'ReadOnly':
emit('from_server', {'cmd': 'model_load_status', 'data': "Loading {}".format(vars.model)}, broadcast=True)
#Have to add a sleep so the server will send the emit for some reason
time.sleep(0.1)
if gpu_layers is not None: if gpu_layers is not None:
args.breakmodel_gpulayers = gpu_layers args.breakmodel_gpulayers = gpu_layers
@ -1254,7 +1265,7 @@ def load_model(use_gpu=True, key='', gpu_layers=None, initial_load=False):
else: else:
num_tensors = len(device_map) num_tensors = len(device_map)
print(flush=True) print(flush=True)
utils.bar = tqdm(total=num_tensors, desc="Loading model tensors") utils.bar = tqdm(total=num_tensors, desc="Loading model tensors", file=Send_to_socketio())
with zipfile.ZipFile(f, "r") as z: with zipfile.ZipFile(f, "r") as z:
try: try:
@ -1871,6 +1882,9 @@ def load_model(use_gpu=True, key='', gpu_layers=None, initial_load=False):
final_startup() final_startup()
if not initial_load: if not initial_load:
set_aibusy(False) set_aibusy(False)
print("Sending model window close")
emit('from_server', {'cmd': 'hide_model_name'}, broadcast=True)
time.sleep(0.1)
# Set up Flask routes # Set up Flask routes

3
static/application.js
View File

@ -2541,10 +2541,11 @@ $(document).ready(function(){
$("#showmodelnamecontainer").removeClass("hidden"); $("#showmodelnamecontainer").removeClass("hidden");
} else if(msg.cmd == 'hide_model_name') { } else if(msg.cmd == 'hide_model_name') {
$("#showmodelnamecontainer").addClass("hidden"); $("#showmodelnamecontainer").addClass("hidden");
//console.log("Closing window");
} else if(msg.cmd == 'model_load_status') { } else if(msg.cmd == 'model_load_status') {
$("#showmodelnamecontent").html("<div class=\"flex\"><div class=\"loadlistpadding\"></div><div class=\"loadlistitem\" style='align: left'>" + msg.data + "</div></div>"); $("#showmodelnamecontent").html("<div class=\"flex\"><div class=\"loadlistpadding\"></div><div class=\"loadlistitem\" style='align: left'>" + msg.data + "</div></div>");
$("#showmodelnamecontainer").removeClass("hidden"); $("#showmodelnamecontainer").removeClass("hidden");
console.log(msg.data); //console.log(msg.data);
} else if(msg.cmd == 'oai_engines') { } else if(msg.cmd == 'oai_engines') {
RemoveAllButFirstOption($("#oaimodel")[0]); RemoveAllButFirstOption($("#oaimodel")[0]);
for (const engine of msg.data) { for (const engine of msg.data) {