Closed waitrouz closed 1 year ago
Hey @waitrouz Sorry, we couldn't get a working deployment instance.
Check your deploy.yml again. Also add to README.md a video demonstration of a working deployment and a link to the main project repository. Thank you!
Hey @waitrouz Sorry, we couldn't get a working deployment instance.
Check your deploy.yml again. Also add to README.md a video demonstration of a working deployment and a link to the main project repository. Thank you!
I think it's an providers problem.
Just successfully deployed this app (after a couple of tries). Hard to find providers (set cpu 4, memory 8, storage 16). OSEQ - 264665. Provider - provider.aitinkerer.com
If app deployed - clear link will give you "502 Bad Gateway", just need to add forwarded port (5002) to url to access to app.
@waitrouz
Perhaps it depends on the type of GPU?
You can use the nvidia-smi
command to query the type on a running deployment, and then specify the type of hardware you want in the attributes.
Also, in yml files, you use different custom images ( ghcr.io/coqui-ai/tts-cpu
and ghcr.io/coqui-ai/tts
) for deployments, but the Dockerfile belongs to only one.
Submit the missing Dockerfile so we can make sure it's secure.
@waitrouz Perhaps it depends on the type of GPU? You can use the
nvidia-smi
command to query the type on a running deployment, and then specify the type of hardware you want in the attributes.Also, in yml files, you use different custom images (
ghcr.io/coqui-ai/tts-cpu
andghcr.io/coqui-ai/tts
) for deployments, but the Dockerfile belongs to only one. Submit the missing Dockerfile so we can make sure it's secure.
This app does not require high-end hardware so problem is not in GPU (drivers maybe).
I left only one yaml file (builded image from Dockerfile of this PR) with comments.
Also again checked testnet providers with this app:
@waitrouz it's work. Modify the root README.md ( https://github.com/akash-network/awesome-akash/blob/master/README.md ) of the awesome-akash repository to include a link to your deployment and you're ready to merge.
Ready to merge @brewsterdrinkwater !
Source: https://github.com/coqui-ai/TTS Youtube video with deploying: https://youtu.be/pRBYgL2Se-o