https://github.com/bluebirdback/ocr-by-gpt-4
Create a script using GPT-4 that can perform Optical Character Recognition (OCR).
https://github.com/bluebirdback/ocr-by-gpt-4
Last synced: over 1 year ago
JSON representation
Create a script using GPT-4 that can perform Optical Character Recognition (OCR).
- Host: GitHub
- URL: https://github.com/bluebirdback/ocr-by-gpt-4
- Owner: BlueBirdBack
- License: mit
- Created: 2023-03-31T00:23:08.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2023-04-19T09:54:06.000Z (about 3 years ago)
- Last Synced: 2025-01-13T14:52:34.915Z (over 1 year ago)
- Language: Python
- Size: 602 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# OCR Script Written by GPT-4
A Python script written by GPT-4 that can perform Optical Character Recognition (OCR) for both English and Chinese characters.
## Getting started
1. Install requirments
```shell
$ pip install requirements.txt
```
2. Install Tesseract
Please refer to https://github.com/UB-Mannheim/tesseract/wiki and https://tesseract-ocr.github.io/tessdoc/Installation.html.
Binaries are available from:
- [Ubuntu - tesseract-ocr-devel PPA](https://launchpad.net/~alex-p/+archive/ubuntu/tesseract-ocr-devel)
- [Debian - notesalexp.org](https://notesalexp.org/tesseract-ocr/#tesseract_5.x)
- [Windows - Tesseract at UB Mannheim](https://github.com/UB-Mannheim/tesseract/wiki)
3. Run the code and Enjoy
```shell
$ python ocr.py IMAGE_PATH [OUTPUT_FILE]
```
## Examples
### [Codepup 3.0 - AI Developer](https://twitter.com/SullyOmarr/status/1648444219932147717)
OCR
```
Tips :
Provide as much detaitLL as to the project goalLs as possibtle in your first message after prompting .
When continuing code，prime it with an unambiLguous snitip of code near where it cut off. ExampLe: Next from "”
{handteLogin}y />”
If CodePup tries to Leave the rest of the coding to you at any poitint，regenerate and stop it tmmediatelLy .
Then，put in your message “Continue with impLementing.."”and state the next part of the appLication to
CompLete .
To avoid issues with conversatiton token Length，get the core of the software comptLeted. Then ask for
additionatL features that are reLated. ExamptLe: "Now add user _ authentication and Login page for admin area.”
Once that is coded，go back to thits message and edit it so it requests a different bit of the rematning
functitions. This keeps the core in-memory and generates compatibtLe functions .
Use ptLain and simplLe Language after describing the project. Consider asking GPT to create deveLopment plLan
from your text .
Define MLLestones for CodePup to foLLow.
Remove the finaL 2 instructions if you don't want it's output to say CodePup .
Major changes: SsLight change of instructions，removalL of needtLess redundant words，condensed further. Now ontLy
145 tokens or 106 without “CodePup ' .
```
## Acknowledgements
- GPT-4 (ChatGPT)