search

ninarina12 / phononDoS_tutorial

Code repository for a tutorial based on the "Direct prediction of phonon density of states with Euclidean neural networks"
MIT License
27 stars 6 forks source link

data download #2

Open yayamy opened 8 months ago

yayamy commented 8 months ago

Hello, how do you download your data.csv? Thanks a lot.

ninarina12 commented 8 months ago

Hi, if you're looking to just pull the data.csv file from the repo, you should be able to do so by clicking on it and selecting the icon to "Download raw file" on the top right. It will also be downloaded if you're cloning the whole repository as described in the README. Hope this helps!

yayamy commented 8 months ago

Hi, if you're looking to just pull the data.csv file from the repo, you should be able to do so by clicking on it and selecting the icon to "Download raw file" on the top right. It will also be downloaded if you're cloning the whole repository as described in the README. Hope this helps!

Thank you for your response. There might be a misunderstanding of my intention. What I meant to ask is: Where did you download your data.csv from? Because the the data in the 'structure' column downloaded using the Material Project API seemed a bit different from yours. yours mine

ninarina12 commented 8 months ago

Hi, sorry for the confusion. The dataset was downloaded directly from the figshare repository provided in the publication of the dataset (https://doi.org/10.6084/m9.figshare.c.3938023.v1), see README for full citation. Note that our structures are saved from ase Atoms objects rather than pymatgen Structure objects, in case that makes a difference. However, if you're finding that the structures are actually non-equivalent, my best guess would be there may have been some updates on Materials Project since the creation of this dataset, so it might be better to use the new, updated versions you're getting. Let me know if you have any other questions or issues.