File size: 918 Bytes
97835c2 21690c9 3955fe2 97835c2 ff0aea3 2c2975c 97835c2 21690c9 fde33e1 2c2975c fde33e1 2c2975c fde33e1 de6be74 97835c2 fde33e1 |
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 |
import gradio as gr
import torch
import spaces
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("ping98k/typhoon-7b-rag-instruct-th")
model = AutoModelForCausalLM.from_pretrained("ping98k/typhoon-7b-rag-instruct-th")
@spaces.GPU(duration=120)
def response(instruction, history, inputText):
inp = f"""### Instruction:
{instruction}
### Input:
=======START OF DOCUMENT=======
{inputText}
=======END OF DOCUMENT=======
### Response:"""
input_ids = tokenizer(inp, return_tensors='pt').to("cuda")
beam_output = model.generate(**input_ids)
outputText = tokenizer.decode(beam_output[0], skip_special_token=True)
#output = output.replace(inp,"").replace("<s>","").replace("</s>","")
return outputText
gr.ChatInterface(
response,
additional_inputs=[
gr.Textbox("You are helpful AI.", label="Input Text"),
],
).launch() |