parity_utils

catlas.parity.parity_utils.get_general_plot(df: pandas.core.frame.DataFrame, output_path: str, number_steps, energy_key1='DFT_energy', energy_key2='ML_energy') → dict

Creates the pdf parity plot for all smiles and returns a dictionary summarizing plot results.

catlas.parity.parity_utils.get_model_id(checkpoint_path: str) → str

get the npz path for a specific checkpoint file

catlas.parity.parity_utils.get_parity_upfront(config, run_id)

Get parity plot to cover intended scope of work in catlas.

Parameters

• config – a dictionary loaded from a catlas input yaml

• run_id – name of the output folder

catlas.parity.parity_utils.get_specific_smile_plot(smile: str, df: pandas.core.frame.DataFrame, output_path: str, number_steps, energy_key1='DFT_energy', energy_key2='ML_energy') → dict

Creates the pdf parity plot for a given smile and returns a dictionary summarizing plot results

catlas.parity.parity_utils.make_parity_plots(df_filtered, config, output_path, number_steps_all)

Makes the general and smile specific parity plots for a given configuration.

catlas.parity.parity_utils.make_subplot(subplot, df, name, number_steps, energy_key1, energy_key2) → dict

Helper function for larger plot generation. Processes each subplot.

catlas.parity.parity_utils.update_info(info_dict: dict, name: str, info_to_add: dict) → dict

Helper function for summary dictionary generation. Updates the dictionary for each split.