QC Fitting Datasets for OpenFF SMIRNOFF Sage 2.3.0

• OpenFF-SMIRNOFF-Sage-2.3.0_optimization_view.sqlite : QCFractal view of optimization dataset 469
• OpenFF-SMIRNOFF-Sage-2.3.0_torsiondrive_view.sqlite : QCFractal view of torsiondrive dataset 470
• docker_handle_dataset_views.tar.gz : Compressed docker image providing a notebook ready to view and handle the data
• README.txt on how to launch the docker image and access the data

Instructions for Dataset Handling

This data is stored a an sqlite file the is directly readable with QCFractal as a dataset view. Because sqlite is a general format, this data may also be accessed in that way, where compressed contents are python picked json strings. To make viewing and handling the data straighforward, a docker image wiith a jupyter notebook entry has been provided.

Load the docker image with:

$ docker load -i docker_handle_dataset_views.tar.gz

A user can run the docker image to spawn the jupyter notebook.

$ mkdir views; mv *sqlite views/
$ mkdir outputs
$ docker run -p 8888:8888 -v ./views:/workspace/views -v ./outputs:/workspace/outputs docker_handle_dataset_views

The -p flag exposed the port 8888 inside the docker image to the port by the same name externally.
The -v flag exposes a directory (in this case ./views, so put your dataset views there) to a directory inside the docker image so that the jupyter notebook and access them. 
The ./outputs directory provides another shared directory that can be useful to pass output files.
If using a M* MAC, the flag --platform=linux/amd64 could be required.

Entering the URL that starts with http://127.0.0.1:8888... in a internet browser should lead to a jupyterlab instance.