https://github.com/gbaeke/nasnet-go
Web app to classify images with nasnetmobile
https://github.com/gbaeke/nasnet-go
ai go golang ml nasnet tensorflow
Last synced: 4 months ago
JSON representation
Web app to classify images with nasnetmobile
- Host: GitHub
- URL: https://github.com/gbaeke/nasnet-go
- Owner: gbaeke
- Created: 2019-02-21T20:55:50.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2021-05-04T21:32:05.000Z (almost 4 years ago)
- Last Synced: 2024-12-12T18:02:44.398Z (5 months ago)
- Topics: ai, go, golang, ml, nasnet, tensorflow
- Language: Go
- Homepage:
- Size: 37.9 MB
- Stars: 4
- Watchers: 2
- Forks: 1
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# nasnet-go
Web app to classify images with nasnetmobile
Use the Docker image gbaeke/nasnet:1.0.5 to try it:
docker run -p 80:9090 -d gbaeke/nasnet:1.0.5
Afterwards, point a browser at http://localhost and try it out
If you want to build and run the app locally, you will need Linux or MacOS. Check the Dockerfile for prerequisites like the TensorFlow C API.
Note: to enable SSL set environment variable ssl equal to true (lowercase) and hostname to the hostname for the certificate. Make sure that the hostname resolves to the container on port 80! The code in main.go uses Let's Encrypt staging CAs
Note: if you deploy to Azure Container Instances, remember that the certificate is stored in ephemeral storage. You might hit Let's Encrypt rate limits if you switch to the production CAs and create the container many times. One solution is to mount an Azure File Share on /root/.local/share/certmagic
Note: with the release of Google Cloud Run, the code was updated to use the PORT environment variable that Cloud Run injects.