| gr.HTML("<h1><center>Nougat: Neural Optical Understanding for Academic Documents<center><h1>") | |
| gr.HTML("<h3><center>Lukas Blecher et al. <a href='https://arxiv.org/pdf/2308.13418.pdf' target='_blank'>Paper</a>, <a href='https://facebookresearch.github.io/nougat/'>Project</a><center></h3>") | |
| with gr.Row(): | |
| mkd = gr.Markdown('<h4><center>Upload a PDF</center></h4>', scale=1) | |
| mkd = gr.Markdown('<h4><center><i>OR</i></center></h4>', scale=1) | |
| mkd = gr.Markdown('<h4><center>Provide a PDF link</center></h4>', scale=1) | |
| with gr.Row(equal_height=True): | |
| pdf_file = gr.File(label='PDF📃', file_count='single', scale=1) | |
| pdf_link = gr.Textbox(placeholder='Enter an Arxiv link here', label='PDF link🔗🌐', scale=1) | |
| with gr.Row(): | |
| btn = gr.Button('Run NOUGAT🍫') | |
| clr = gr.Button('Clear🚿') | |
| output_headline = gr.Markdown("<h3>PDF converted to markup language through Nougat-OCR👇:</h3>") | |
| parsed_output = gr.Markdown(elem_id='mkd', value='📃🔤OCR Output') | |
| mmd_file_download = gr.File(label='Download .mmd file', interactive=False) | |
| def handle_predict(pdf_file, pdf_link): | |
| content, mmd_file_path = predict(pdf_file, pdf_link) | |
| return gr.update(value=content), mmd_file_path | |
| btn.click(handle_predict, [pdf_file, pdf_link], [parsed_output, mmd_file_download]) | |
| clr.click(lambda: (gr.update(value=None), | |
| gr.update(value=None), | |
| gr.update(value=None)), | |
| [], | |
| [pdf_file, pdf_link, parsed_output, mmd_file_download]) | |
| gr.Examples( | |
| [["input/nougat.pdf", ""], [None, "https://arxiv.org/pdf/2308.08316.pdf"]], | |
| inputs=[pdf_file, pdf_link], | |
| outputs=parsed_output, | |
| fn=process_example, | |
| cache_examples=True, | |
| label='Click on any Examples below to get Nougat OCR results quickly:' | |
| ) | |
| demo.queue() | |
| demo.launch(debug=True) | |