File size: 1,528 Bytes
2340542 ab948b0 204123c 17eb9c7 388efe5 17eb9c7 2e73e37 189e034 17eb9c7 388efe5 17eb9c7 38f9c6c 388efe5 17eb9c7 2e73e37 17eb9c7 | 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 | import gradio as gr
import torch
from transformers import GPT2Tokenizer, GPT2LMHeadModel
checkpoint = "gpt2"
tokenizer = GPT2Tokenizer.from_pretrained(checkpoint)
# Load the fine-tuned model and tokenizer
my_model = GPT2LMHeadModel.from_pretrained("jeevana/GenerativeQnASystem")
my_tokenizer = GPT2Tokenizer.from_pretrained("jeevana/GenerativeQnASystem")
def generate_response(model, tokenizer, prompt):
input_ids = tokenizer.encode(prompt, return_tensors="pt",truncation=True, max_length=1000)
# Create the attention mask and pad token id
attention_mask = torch.ones_like(input_ids)
pad_token_id = tokenizer.eos_token_id
output = model.generate(
input_ids,
max_new_tokens=70,
min_new_tokens = 1,
num_return_sequences=1,
attention_mask=attention_mask,
pad_token_id=pad_token_id
)
qna = tokenizer.decode(output[0], skip_special_tokens=True)
answer = qna[len(prompt)+9: ]
return answer
def generative_qna(input):
response = generate_response(my_model, my_tokenizer, input)
return response
# def generative_qna(input):
# print(input)
# return input
app = gr.Interface(fn=generative_qna, inputs=[gr.Textbox(label="Question", lines=3)],
outputs=[gr.Textbox(label="Answer", lines=6)],
title="Generative QnA System",
description="Generative QnA with GPT2"
)
app.launch(share=True, debug=True)
# gr.load("models/jeevana/GenerativeQnASystem").launch() |