NVIDIA / gpu-rest-engine

A REST API for Caffe using Docker and Go
BSD 3-Clause "New" or "Revised" License
421 stars 94 forks source link

Using DIGITS trained network in rest engine #4

Closed emerysilb closed 8 years ago

emerysilb commented 8 years ago

Hello, I am trying to use a DIGITS trained network in this gpu-rest-engine and I am having some issues setting up the docker container with the pre-trained network. What is the best way for importing the pre-trained network into the rest engine for processing?

Thank you!

flx42 commented 8 years ago

Hello,

First, remove the bundled caffenet model, you won't need it. Remove these lines

And then, you should use a Docker volume to expose the trained model inside the container. For instance:

-v /home/felix/models:/models

Finally, modify the paths used at the end of the Dockerfile to point to the model you want to use, for instance:

/models/mydigitsmodel/deploy.prototxt
emerysilb commented 8 years ago

Thank you so much! This program is awesome!!!