Experiment Factory Experiment
Hi Friend! This is an Experiment (adopted from gewimmer-neuro/reward_learning_js) that is friendly for use in the Experiment Factory. You can run it locally by putting these files in a web server, or use the Experiment Factory to generate a reproducible container. Check out the documentation above for more information, or post an issue if you have any questions.
Build an Experiment Container
These instructions are also available here.
First, create a working directory
mkdir -p /tmp/reward-learning cd /tmp/reward-learning
Then see experiments available
docker run quay.io/vanessa/expfactory-builder list
Generate a container with the reward-learning-task
docker run -v $PWD:/data quay.io/vanessa/expfactory-builder build reward-learning-task
The message will tell you the next step - to build your container! And actually, you would be best off (if you want to share or reproduce this) to add the Dockerfile to a GitHub repository and then have an automated build.
Expfactory Version: 3.16 LOG Recipe written to /data/Dockerfile To build, cd to directory with Dockerfile and: docker build --no-cache -t expfactory/experiments .
/data folder in the container is where you just bound the present working directory,
so our Dockerfile and entrypoint script are actually right here!
$ ls Dockerfile startscript.sh
We could build that as follows:
docker build -t reward-learning .
And then run it on port 80:
mkdir -p /tmp/reward-learning/data docker run -v /tmp/reward-learning/data/:/scif/data -p 80:80 reward-learning start
And then open your browser to 127.0.0.1 to see the interface!
By default, the data is saved to the filesystem where you mounted the local data folder.
And of course see the documentation pages for how to customize the database, and other configuration. If you need to customize the experiment repository cloned from (e.g., if you want to make changes) you can edit this in the Dockerfile:
LABEL EXPERIMENT_reward-learning-task /scif/apps/reward-learning-task WORKDIR /scif/apps RUN expfactory install https://www.github.com/<username>/reward-learning-task