We are working on a more efficient and powerful OCR pipeline. Please stay tuned.
This repo contains a colab notebook that parses pdfs in google drive into text files for further analysis
If you have a lot of PDFs scanned in your google drive and you cannot do NLP analysis on it, you are in the RIGHT place!!!
Our project parses and organizes your text for you; all you need to do is to put your PDFs in a common folder.
- Have a folder of PDFs that you want to run OCR on in your google drive. (ex. PDFs)
- Save this PDF2Page.ipynb colab notebook anywhere in your google drive.
- Open the PDF2Page.ipynb and change the first cell accordingly:
a. Input folder’s google id
b. Input folder’s directory
c. Output folder’s directory (which you name) - Run the entire notebook. When finished, you can find all results in the output folder.
In your output folder, you will find the following:
- A folder for each pdf documents you want to parse, the folder name is the google id of the original pdf.
- Inside each folder, you will find each page as a picture.
- Inside each folder, you will find a page.csv file. It will have page number mapped to the ocr results for each page.
Adam Anderson, Jason Webb, Kenan Jiang, Kevin Gao, Kewei Chen, Max Ziff, Serena Zhang, Shuyao Zhou
Adam Anderson: admndrsn@berkeley.edu
Kenan Jiang: kenanj11@berkeley.edu
Kevin Gao: kevingao26@berkeley.edu