CSV file "competition_GS-PES.csv"
This csv defines which data in
Small_Molecules_1 dataset to be calculated in the competition.
Each row corresponds to the data point whose the ground state energy needs to be calculated, and also contains information to calculate the score in the competition.
dirname: name of directory in which
filename: name of
molecule_name: name of molecule
distance: bond length of molecule (see here for definition)
fci_energy: exact ground state energy calculated by the full configuration interaction
1st_excited_energy: exact 1st excited state energy in the singlet (spin $S^2=0$) sector calculated by the full configuration interaction
bond_length_opt: optimal bond length in which
fci_energytakes minimum (4 digits accuracy after the decimal point). This value is common among data points for the same molecule.
fci_energy_lowest: ground state energy at
qc_energy(empty): ground state energy calculated by quantum algorithms/methods.
nfev(empty, optional): number of function evaluation. The definition varies for each algorithms/methods.
How to use csv and join competition
Participants of the competition fill
qc_energy columns and
nfev (optional) of the csv.
A Python script
process_competition_GS_PES.py in this directory processes the csv and generale the result summaries.
It will generate a figure of the potential energy surface and a detailed data table for each molecule. Moreover, it will create
summary_GS-PES.csv which summarizes results for all molecules and
record_GS-PES.csv which contains a one-line summary of the results.
For further information of the script, please see the docstring of it.