Spaces:
Sleeping
Sleeping
File size: 2,920 Bytes
7b4f776 23a0568 552d9ba 23a0568 7b4f776 23a0568 7b4f776 23a0568 7b4f776 23a0568 7b4f776 23a0568 7b4f776 23a0568 7b4f776 23a0568 7b4f776 23a0568 7b4f776 |
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 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 |
import gradio as gr
import pytube
from youtube_transcript_api import YouTubeTranscriptApi as yt
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import os
from langchain import PromptTemplate
from langchain import LLMChain
from langchain_together import Together
# Set the API key with double quotes
os.environ['TOGETHER_API_KEY'] = "d88cb7414e4039a84d2ed63f1b47daaaa4230c4c53a422045d8a30a9a3bc87d8"
def Summary_BART(text):
checkpoint = "sshleifer/distilbart-cnn-12-6"
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint)
inputs = tokenizer(text, max_length=1024, truncation=True, return_tensors="pt")
summary_ids = model.generate(inputs["input_ids"])
summary = tokenizer.batch_decode(summary_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)
return summary[0]
def YtToQuizz(link, difficulty_level):
video_id = pytube.extract.video_id(link)
transcript = yt.get_transcript(video_id)
data = ""
for text in transcript:
data += text.get('text')
summary = Summary_BART(data)
mcq_template = """
Give a 10 different multiple-choice question MCQ related to the summary: {summary}
The difficulty level of the question should be: {difficulty_level}
Please provide the following in:
1. Question
2. Correct answer
3. Three plausible incorrect answer options
4. Proper MCQ format
"""
prompt = PromptTemplate(
input_variables=['summary', 'difficulty_level'],
template=mcq_template
)
llama3 = Together(model="meta-llama/Llama-3-70b-chat-hf", max_tokens=2500)
Generated_mcqs = LLMChain(llm=llama3, prompt=prompt)
response = Generated_mcqs.invoke({
"summary": summary,
"difficulty_level": difficulty_level
})
response_text = response['text']
# Extract MCQs, correct answers, and options
questions = re.findall(r'Question: (.*?)\n', response_text)
correct_answers = re.findall(r'Correct answer: (.*?)\n', response_text)
options = re.findall(r'Options: (.*?)\n', response_text)
all_options = [option.split(', ') for option in options]
return questions, all_options, correct_answers
def main(link, difficulty_level):
questions, options, correct_answers = YtToQuizz(link, difficulty_level)
return {
"Questions": questions,
"Options": options,
"Correct Answers": correct_answers
}
iface = gr.Interface(
fn=main,
inputs=[
gr.inputs.Textbox(lines=2, placeholder="Enter YouTube video link"),
gr.inputs.Dropdown(["Easy", "Medium", "Hard"], label="Select difficulty level:")
],
outputs=[
gr.outputs.JSON(label="MCQs Output")
],
title="YouTube Video Subtitle to MCQs Quiz",
description="Generate MCQs from YouTube video subtitles"
)
if __name__ == '__main__':
iface.launch()
|