File size: 2,153 Bytes
c57ccc5 70f1dd6 c57ccc5 70f1dd6 c57ccc5 a13b0da 70f1dd6 |
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 58 59 60 61 62 63 64 65 66 |
import gradio as gr
# Method 1
# from transformers import pipeline
# question_answer = pipeline('question-answering',model = 'distilbert/distilbert-base-cased-distilled-squad')
# def question_answering(context, question):
# output = question_answer({
# 'context': context,
# 'question': question
# })
# return output['answer'], str(output['score'] * 100)
# Method 2
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
import torch
import torch.nn.functional as F
tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-cased-distilled-squad")
model = AutoModelForQuestionAnswering.from_pretrained("distilbert/distilbert-base-cased-distilled-squad")
def question_answering(context, question):
inputs = tokenizer.encode_plus(question, context, return_tensors="pt")
# Get input IDs and attention mask
input_ids = inputs["input_ids"].tolist()[0]
# Perform inference to get the start and end logits
outputs = model(**inputs)
start_logits = outputs.start_logits
end_logits = outputs.end_logits
# Get the most likely beginning and end of answer with the argmax of the logits
start_index = torch.argmax(start_logits)
end_index = torch.argmax(end_logits) + 1
# Apply softmax to get probabilities
start_probs = F.softmax(start_logits, dim=-1)
end_probs = F.softmax(end_logits, dim=-1)
# Convert token IDs of the answer span back to text
answer = tokenizer.convert_tokens_to_string(tokenizer.convert_ids_to_tokens(input_ids[start_index:end_index]))
# Calculate the confidence score
confidence_score = start_probs[0][start_index].item() * end_probs[0][end_index-1].item()
return answer, str(confidence_score * 100)
iface = gr.Interface(
fn = question_answering,
inputs = [
gr.Textbox(label = "Context", placeholder = "Enter your context here..", lines = 5),
gr.Textbox(label = "Question", placeholder = "Enter your question", lines = 2)
],
outputs = [
gr.Textbox(label = "Answer", lines = 2),
gr.Textbox(label = "Accuracy")
]
)
iface.launch() |