Spaces:
Sleeping
Sleeping
File size: 1,966 Bytes
080cf4b c476863 080cf4b 079db35 080cf4b eeb55b5 080cf4b c476863 |
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 |
import gradio as gr
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
from datetime import datetime
print('{}:loading...'.format(datetime.now().strftime('%Y-%m-%d %H:%M:%S')))
tokenizer = AutoTokenizer.from_pretrained('line-corporation/japanese-large-lm-1.7b-instruction-sft', use_fast=False)
model = AutoModelForCausalLM.from_pretrained('line-corporation/japanese-large-lm-1.7b-instruction-sft')
#tokenizer = AutoTokenizer.from_pretrained('line-corporation/japanese-large-lm-3.6b-instruction-sft', use_fast=False)
#model = AutoModelForCausalLM.from_pretrained('line-corporation/japanese-large-lm-3.6b-instruction-sft')
if torch.cuda.is_available():
model.half()
model = model.to('cuda')
generator = pipeline('text-generation', model=model, tokenizer=tokenizer, device=model.device)
print('{}:done.'.format(datetime.now().strftime('%Y-%m-%d %H:%M:%S')))
def generate(input_text, maxlen):
input = f'ユーザー: {input_text}\nシステム: '
output = generator(
input,
max_length=maxlen,
do_sample=True,
temperature=0.7,
top_p=0.9,
top_k=0,
repetition_penalty=1.1,
num_beams=1,
num_return_sequences=1,
pad_token_id=tokenizer.pad_token_id,
bos_token_id=tokenizer.bos_token_id,
eos_token_id=tokenizer.eos_token_id
)
generated_text = output[0]['generated_text'][len(input) + 1:]
return generated_text
with gr.Blocks(title='question answering ja') as app:
gr.Markdown('# Question Answering JA')
chatbot = gr.Chatbot(label='answer')
msg = gr.Textbox(label='question')
maxlen = gr.Slider(minimum=30, maximum=100, value=30, step=1, label='max length')
clear = gr.ClearButton([msg, chatbot])
def respond(message, maxlen, chat_history):
if message == '':
return '', chat_history
bot_message = generate(message, maxlen)
chat_history.append((message, bot_message))
return '', chat_history
msg.submit(respond, [msg, maxlen, chatbot], [msg, chatbot], concurrency_limit=20)
app.launch()
|