https://github.com/jenspapenhagen/image-captioning

playing around with image-captioning model and make it prod ready
https://github.com/jenspapenhagen/image-captioning

Last synced: 1 day ago
JSON representation

playing around with image-captioning model and make it prod ready

• Host: GitHub
• URL: https://github.com/jenspapenhagen/image-captioning
• Owner: jenspapenhagen
• Created: 2023-03-06T16:05:01.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2024-11-18T17:52:27.000Z (2 days ago)
• Last Synced: 2024-11-18T18:53:04.253Z (2 days ago)
• Language: Python
• Homepage:
• Size: 201 KB
• Stars: 0
• Watchers: 4
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# image-captioning

playing around with image-captioning model and try to make it prod ready.

1. add an python flask http endpoint

2. move this endpoint into a docker image and mount the model and images from the host

3. try to make flask server prod save. with waitress server

    1. adding CORS

    2. adding Prometheus Metrics

4. loading the model in an extzra service for better seperation

5. testing other servicer than waitress

6.

Infos:

the model is: [https://huggingface.co/nlpconnect/vit-gpt2-image-captioning](nlpconnect/vit-gpt2-image-captioning)

this need be downloaded separate and place in the folder modle/transformers/*