I am using Python to read many PDFs, and they are quite large (some have 40 pages, others 3000). Therefore, I need some optimization. As such, these PDFs have tabulated data and everything is well-structured. From these tables, I need to match a record and extract a row from each PDF, but since there are so many, my kernel crashes... I was using Pdfplumber and also tried with LlamaReadMarkdown. Any suggestions? Here is my code with pdfplumber.open(pdf_path) as pdf: for page in pdf.pages[1:]: tables = page.extract_table() if tables: df = pd.DataFrame(tables[1:], columns=tables[0]) for index, row in df.iterrows(): if rut in row.values: return df.loc[[index]] return None Im thinking to process it by batch but it will be viable? Continue reading...
