Some Jupyter / Pandas notebooks I'm playing around with. To use, first install (Docker)[https://docs.docker.com/engine/installation/] on your machine.
To start a datascience
environment, run $ source scripts/ds_start.sh
This comes with R
, Python
, and Julia
kernels, all with "batteries included",
with the exception of TensorFlow
. This will serve the notebooks at http://localhost:8888
To start a tensorflow
environment, run $ source scripts/tf_start.sh
. This starts
TensorFlow and TensorBoard. This will serve the notebooks at http://localhost:8889
.
You can run both simulataneously, and they will share the working directory. It's not a good idea to open the same notebook in both simulataneously, but you can pass data around over disk.
Once you are done, you can use docker stop $DS_CONTAINER
or docker stop $TF_CONTAINER
to stop the docker process.