| import gradio as gr | |
| from langchain.document_loaders import PDFMinerLoader, PyMuPDFLoader | |
| from langchain.text_splitter import CharacterTextSplitter | |
| def extract_text(pdf_file): | |
| # Load a document | |
| loader = PDFMinerLoader("cereal.pdf") | |
| doc = loader.load() | |
| text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0) | |
| texts = text_splitter.split_documents(doc) | |
| texts = [i.page_content for i in texts] | |
| return texts[0] | |
| iface = gr.Interface( | |
| fn=extract_text, | |
| inputs=gr.File(type="file", label="Upload PDF"), | |
| outputs="text" | |
| ) | |
| iface.launch() | |