# Filter the jobs that were unsuccessful
df_m = df_m[[not i for i in pd.isna(df_m["elec_energy"].tolist())]]
df_m = df_m[df_m["job_type"] == "surface_coverage_energy"]
cols_to_keep = [
'facet',
'job_type',
'layers',
'surface_type',
'elec_energy',
'atoms_object',
'bulk_system',
'coverage_type',
]
df_m = drop_columns(df=df_m, columns=cols_to_keep, keep_or_drop="keep")
# #############################################################################
# Read bulk data ##############################################################
with open(oer_bulk_structures_path, "rb") as fle:
bulk_data = pickle.load(fle)
# + attributes={"classes": [], "id": "", "n": "5"}
path_i = os.path.join(os.environ["PROJ_irox"],
"workflow/07_bulk_pourbaix/01_pourbaix_scripts",
"out_data/bulk_pourb_transitions.pickle")
with open(path_i, "rb") as fle:
pourb_trans = pickle.load(fle)
pourb_trans
# -
# df_ccf = DF_dict["df_ccf"]
df_dij = DF_dict["df_dij"]
# ids_to_discard__too_many_atoms = DF_dict["ids_to_discard__too_many_atoms"]
# df_dft_final_final = DF_dict["df_dft_final_final"]
df_prototype_dft_path = DF_dict["df_prototype_dft"]
df_prototype_static_path = DF_dict["df_prototype_static"]
# +
from misc_modules.pandas_methods import drop_columns
df_prototype_dft_path = drop_columns(
df=df_prototype_dft_path,
columns=[
# 'p_name',
'structure_name',
# 'spacegroup',
'wyckoffs',
'species',
],
keep_or_drop="drop")
df_bulk_dft = drop_columns(
df=df_bulk_dft,
columns=[
"atoms",
"form_e_chris",
"id",
"id_old",
"path",
"volume",
# "",
df_m.loc[df_m["coverage_type"] == "O-2_OH-2", "coverage_type"] = "h_covered"
# #############################################################################
# Drop unnecessary columns from dataframe #####################################
drop_cols = [
'bulk_system',
'facet',
'adsorbate',
'coverage_type',
'ads_e',
'elec_energy',
'surface_type',
"ooh_direction",
]
df_m = drop_columns(df=df_m, columns=drop_cols, keep_or_drop="keep")
# Resetting index to have unique id to pass to OXR module
df_m = df_m.reset_index()
# __|
# -
# # | - Fitting only O-covered data
# +
df_o = df_m[df_m["coverage_type"] == "o_covered"]
df_m_tmp = df_o
ORR_PLT = ORR_Free_E_Plot(
free_energy_df=None,
state_title="adsorbate",
df = drop_columns(df=df, keep_or_drop="keep",
columns=[
"bulk_system",
"facet",
"adsorbate",
"coverage_type",
# "ooh_direction",
"ads_e",
# "elec_energy",
# "total_magmom",
# "abs_magmom",
# "path_short",
# "name_i",
# "max_force",
# "sum_force",
# "elem_num_dict",
# "incar_parsed",
# "init_atoms",
# "atoms_object",
# "N_atoms",
# "dipole_correction",
# "path",
# "name_i_2",
# "name_i_3",
# "priority",
"surface_type",
],
)
data_list = []
grouped = df_iro2_long.groupby(["id"])
for name, group in grouped:
# df_succ = group[group["job_state"] == "SUCCEEDED"]
df_succ = group[group["completed"] == True]
isif_2_done = 2 in df_succ["isif"].tolist()
if len(df_succ) > 0 and isif_2_done:
latest_succ_rev = df_succ.sort_values("revision").iloc[-1]
data_list.append(latest_succ_rev)
df_iro2 = pd.DataFrame(data_list)
# Droping all unnecessary columns
df_iro2 = drop_columns(df=df_iro2,
columns=["atoms", "path"],
keep_or_drop="keep")
# Adding stoich column
df_iro2["stoich"] = "AB2"
# -
# # IrO3 Bulk Data
# +
# #############################################################################
# Parsing NERSC IrO3 DFT Data
path_i = os.path.join(os.environ["PROJ_DATA"],
"04_IrOx_surfaces_OER/ml_bulk_irox_dft/iro3",
"df_dict_nersc.pickle")
# "df_dict.pickle")
# # Main Loop
# +
from misc_modules.pandas_methods import drop_columns
# print(list(df_jobs_anal.columns))
# ['job_id_max', 'timed_out', 'completed', 'brmix_issue', 'job_understandable', 'decision', 'dft_params_new', 'job_completely_done']
cols_to_keep = [
"job_id_max",
"job_completely_done",
]
df_jobs_anal_i = drop_columns(
df=df_jobs_anal,
columns=cols_to_keep,
keep_or_drop="keep",
)
# +
# assert False
# +
index_keys = list(df_jobs_anal_i.index.names)
# #########################################################
data_dict_list = []
# #########################################################
for index_i, row_i in df_jobs_anal_i.iterrows():
# #####################################################
data_dict_i = dict()
# #########################################################
dir_i = os.path.join(os.environ["PROJ_irox_oer"],
"workflow/process_bulk_dft/standardize_bulks", "out_data")
file_name_i = os.path.join(dir_i, "df_dft_stan.pickle")
with open(file_name_i, "rb") as fle:
df_dft_stan = pickle.load(fle)
# +
# #####################################################
df_dft = drop_columns(df=df_dft,
columns=[
"atoms",
"form_e_chris",
"id_old",
"path",
"id",
"source",
"energy",
],
keep_or_drop="drop")
df_dft = df_dft.sort_values("dH")
# -
# Create new atoms column from `atoms_dict`
df_dft["atoms"] = df_dft.index.map(atoms_dict)
# +
# df_dft.drop(columns=["num_atoms", "", ])