foundry_osdb_v1-1 / README.md
blaiszik's picture
Upload README.md with huggingface_hub
9d86cec verified
metadata
license: cc-by-4.0
task_categories:
  - tabular-regression
  - tabular-classification
tags:
  - materials-science
  - chemistry
  - foundry-ml
  - scientific-data
size_categories:
  - 1K<n<10K

Data for: Ab initio control of zeolite synthesis and intergrowth with high-throughput simulations

Dataset Information

Fields

Field Role Description Units
crystal_id input unique identifier associated with each pose. It is
Zeolite input IZA code of the zeolite
SMILES input SMILES string of the guest docked in the zeolite
InchiKey input InchiKey of the guest docked in the zeolite
Ligand formula input formula of one molecular guest
Loading input number of OSDAs per unit cell in the calculated po
Binding (SiO2) target binding energy between the host and the guest, cal kJ/mol
Binding (OSDA) target binding energy between the host and the guest, cal
Directivity (SiO2) target binding energy between the host and the guest, usi kJ/mol
Competition (SiO2) target competition energy between different hosts for a g kJ/mol
Competition (OSDA) target competition energy between different hosts for a g kJ/mol
Templating target templating energy at 400 K, as calculated in the p kJ/mol
SCScore input Synthetic Complexity Score, as proposed by Coley e kJ/mol
Volume (Angstrom3) input volume of the OSDA, given in Angstrom^3. Angstrom^3
Axis 1 (Angstrom) input first principal component of the OSDA, given in An Angstrom
Axis 2 (Angstrom) input second principal component of the OSDA, given in A Angstrom
In literature? input If the pair is known in the literature, the value kJ/mol
lattice input lattice matrix of the crystal
nxyz input tuple containing the atomic number and the (x, y,

Splits

  • train: train

Usage

With Foundry-ML (recommended for materials science workflows)

from foundry import Foundry

f = Foundry()
dataset = f.get_dataset("10.18126/c5z9-zej7")
X, y = dataset.get_as_dict()['train']

With HuggingFace Datasets

from datasets import load_dataset

dataset = load_dataset("foundry_osdb_v1.1")

Citation

@misc{https://doi.org/10.18126/c5z9-zej7
doi = {10.18126/c5z9-zej7}
url = {https://doi.org/10.18126/c5z9-zej7}
author = {Schwalbe-Koda, Daniel and Gómez-Bombarelli, Rafael}
title = {Data for: Ab initio control of zeolite synthesis and intergrowth with high-throughput simulations}
keywords = {machine learning, foundry, zeolite, database}
publisher = {Materials Data Facility}
year = {root=2021}}

License

CC-BY 4.0


This dataset was exported from Foundry-ML, a platform for materials science datasets.