1/1
JARVIS-ML-CFID-descriptors and material properties
Version 2 2020-04-29, 22:50
Version 1 2018-07-26, 22:08
dataset
posted on 2020-04-29, 22:50 authored by Kamal ChoudharyKamal ChoudharyClassical force-field inspired descriptors (CFID) for more than 35000 materials and their material properties such as bandgap, formation energies, modulus of elasticity etc.
See JARVIS-ML:
https://jarvis.nist.gov/
########################################
Example:
import os,json
import pandas as pd
f = open(os.path.join(os.getcwd(),'jml_3d-4-26-2020.json'),'r')
dataml=json.load(f)
f.close()
df=pd.DataFrame(dataml)
typical_data_ranges = {'formation_energy_peratom': [-5, 5],
'optb88vdw_bandgap': [0, 10],
'mbj_bandgap': [0, 10],
'bulk_modulus_kv': [0, 250],
'shear_modulus_gv': [0, 250],
'epsx': [0, 60],
'epsy': [0, 60],
'epsz': [0, 60],
'mepsx': [0, 60],
'mepsy': [0, 60],
'mepsz': [0, 60],
'n-Seebeck': [-600, 10],
'n-powerfact': [0, 5000],
'p-Seebeck': [-10, 600],
'p-powerfact': [0, 5000],
'slme': [0, 40],
'spillage': [0, 4],
'encut': [0, 2000],
'kpoint_length_unit': [0, 200],
'dfpt_piezo_max_dielectric': [0, 100],
'dfpt_piezo_max_dij': [0, 3000],
'dfpt_piezo_max_eij': [0, 10],
'ehull': [0, 1],
'electron_avg_effective_masses_300K': [0, 3],
'hole_avg_effective_masses_300K': [0, 3],
'exfoliation_energy': [0, 1000],
'magmom_oszicar': [0, 10],
'max_ir_mode': [0, 4000],
'total_energy_per_atom': [-10, 3]}import numpy as np
%matplotlib inline
import matplotlib.pyplot as plt
val=np.array(df['formation_energy_peratom'].replace('na',np.nan).dropna().values,dtype='float')
plt.hist(val,bins=np.arange(-4,4,.5))
#########################################
For more details about using the dataset, use the jupyter-notebooks:
