test_2 / test.py
astro21's picture
Create test.py
c2b575f
# import gradio as gr
# from transformers import pipeline
# import os
# import pandas as pd
# # Load the text summarization pipeline
# summarizer = pipeline("summarization", model="astro21/bart-cls")
# chunk_counter = 0
# def summarize_text(input_text):
# global chunk_counter
# chunk_counter = 0
# max_chunk_size = 1024
# chunks = [input_text[i:i + max_chunk_size] for i in range(0, len(input_text), max_chunk_size)]
# summarized_chunks = []
# chunk_lengths = []
# summarized_chunks_only = []
# for chunk in chunks:
# chunk_counter += 1
# summarized_chunk = summarizer(chunk, max_length=128, min_length=64, do_sample=False)[0]['summary_text']
# summarized_chunks.append(f"Chunk {chunk_counter}:\n{summarized_chunk}")
# summarized_chunks_only.append(summarized_chunk)
# chunk_lengths.append(len(chunk))
# summarized_text = "\n".join(summarized_chunks)
# summarized_text_only = "\n".join(summarized_chunks_only)
# # Save the merged summary to a file
# with open("summarized.txt", "w") as output_file:
# output_file.write(summarized_text_only)
# chunk_df = pd.DataFrame({'Chunk Number': range(1, chunk_counter + 1), 'Chunk Length': chunk_lengths})
# return summarized_text, chunk_df, "summarized.txt"
# def read_file(file):
# print(file[0].name)
# with open(file[0].name, 'r') as file_:
# content = file_.read()
# return content
# def summarize_text_file(file):
# if file is not None:
# content = read_file(file)
# return summarize_text(content)
# input_type = gr.inputs.File("text")
# # Name the outputs using the label parameter and provide a download option
# demo = gr.Interface(fn=summarize_text_file, inputs=input_type,
# outputs=[gr.Textbox(label="Summarized Text"),
# gr.Dataframe(label="Chunk Information", type="pandas"),
# gr.File(label="Download Summarized Text", type="file", live=False)],
# title = "Text Summarization",
# description = "Summarize text using BART",
# theme = "huggingface",
# allow_flagging="never",
# live=True)
# demo.launch(share=True)