Currently we copy the sorting logic of the [emmet](https://github.com/materialsproject/emmet) package of Materials Project to pick the right tasks for a given material. However, we might want to use directly the package for compatibility with feature updates. This will need to be thought through as it might not be easy to do given that the data we use is downloaded from AWS OpenData rather than the API.
Currently we copy the sorting logic of the
[emmet](https://github.com/materialsproject/emmet)package of Materials Project to pick the right tasks for a given material. However, we might want to use directly the package for compatibility with feature updates. This will need to be thought through as it might not be easy to do given that the data we use is downloaded from AWS OpenData rather than the API.