Closed piercus closed 6 years ago
Hello
This package looks very promising, and i would like to suggest more plugins (opencv, canvas) to implement.
Anyway, i'd like to first make tensorflow work and i'm failing now on the deploy step
It was failing with error
$ sls deploy ... Error: ENOENT: no such file or directory, open '.ephemeral/lib/tensorflow-1.4.0.zip'
I found that when running docker-composer on his own i am facing a configuration error
The Docker Engine version is less than the minimum required by Compose. Your current project requires a Docker Engine of version 1.13.0 or greater.
So, i have configuration issue on my side, but the library is not raising the right error, (i have spent 30 min in the code to find the root cause).
I think shelljs error are not caught. After digging a little bit inside the code, i think problem comes from the shelljs exec :
const build = shell.exec('docker-compose build', { async: true }); return new Promise((resolve, reject) => { build.on('error', err => reject(err)); build.on('close', () => resolve()); });
I don't think this will catch any error occuring on the 'docker-composer' command
I will work on a PR for this
Hello
This package looks very promising, and i would like to suggest more plugins (opencv, canvas) to implement.
Anyway, i'd like to first make tensorflow work and i'm failing now on the deploy step
It was failing with error
I found that when running docker-composer on his own i am facing a configuration error
So, i have configuration issue on my side, but the library is not raising the right error, (i have spent 30 min in the code to find the root cause).
I think shelljs error are not caught. After digging a little bit inside the code, i think problem comes from the shelljs exec :
I don't think this will catch any error occuring on the 'docker-composer' command
I will work on a PR for this