I would like to use the structure predictor to generate some configurations for Ti(OH)2 and I would like to share these results if possible. Is there any way to either convince the stucture predictor to submit them on your cluster or to submit them on my resources and send them to you.
Absolutely, you can use the Crystal Toolkit to submit a structure into our workflow infrastructure. Drag and drop a CIF file (e. g. from the structure predictor results) into the Crystal Toolkit interface and press “Submit” to do so.
If I wanted to run the job on my end is there any way to do that and then give it to you for databasing?
At present, there is not an official way. However, we’re in the process of reworking our builder and workflow infrastructure to more effectively allow people to run calculations according to our preset parameters and contribute them. The atomate package (https://pythonhosted.org/atomate/), which is essentially a repository of workflows that are used to generate all of the data we have (bandstructure, elastic tensor, piezo, etc.) with our preset parameters is one part of this that’s more or less ready to go, but we’re still in the process of updating the builder infrastructure such that we can seamlessly accept external results.
Also, a new system of user contributions (tentatively called MPContribs) in which users might contribute data that wasn’t explicitly run using our methods (or comes from experiment) but can be attached to our data framework is in development.
We’re devoting a significant portion of our resources to getting this done over the course of the summer, so feel free to check back in around a month or two. I’ll make a note in my calendar to revisit this thread with an update in mid-July, for anyone else who’s interested.
I am still interested in this, my calculations have been sitting at Ready for a long time but have not run. I will just have to run these myself but it seems like a waste to not also be able to submit these for others to use.
Thanks for reaching out. I’ve been informed just a few days ago that unfortunately many of our jobs are failing on the computing resource responsible for these calculations. I still have to look into why this is happening. I hope I can get things up and running again soon. If so, I’ll prioritize your calculations and shepard them through our production to make sure they finish quickly.
sorry for the delay,
just wanted to let you know that MPComplete is running again and your submissions should be picked up very soon.
That is greatly appreciated, if the option to eventually assist with calculations is available though please keep me updated.
Thank you for fixing the jobs so that they would run, but they fizzled at the step where they were to be added to the database it seems. Can someone review this?
thanks for bringing this to our attention. We’re currently preparing for the MP workshop this week. I’ll get back to you as soon as I had the chance to check out your submission when things calmed down again.