from methods import isnotebook
isnotebook_i = isnotebook()
if isnotebook_i:
from tqdm.notebook import tqdm
verbose = True
show_plot = True
else:
from tqdm import tqdm
verbose = False
show_plot = False
# ### Read Data
# +
from methods import get_df_features_targets
df_features_targets = get_df_features_targets()
from methods import get_df_slab
df_slab = get_df_slab()
from methods import get_df_SOAP_AS, get_df_SOAP_MS, get_df_SOAP_ave
df_SOAP_AS = get_df_SOAP_AS()
df_SOAP_MS = get_df_SOAP_MS()
df_SOAP_ave = get_df_SOAP_ave()
# #########################################################
df_i = df_features_targets
# Getting phase > 1 slab ids
df_slab_i = df_slab[df_slab.phase > 1]
phase_2_slab_ids = df_slab_i.slab_id.tolist()
import pickle
import numpy as np
import plotly.graph_objs as go
from sklearn.linear_model import LinearRegression
# #########################################################
from methods import get_df_features_targets
# -
# ### Read Data
# +
df_mine = get_df_features_targets()
# #########################################################
path_i = os.path.join(
os.environ["PROJ_irox_oer"],
"workflow/seoin_irox_data/featurize_data",
"out_data/df_features_targets.pickle")
with open(path_i, "rb") as fle:
df_seoin = pickle.load(fle)
# #########################################################
# +
data = []
elec_or_gibbs = "g"
# Pickling data ###########################################
directory = os.path.join(
os.environ["PROJ_irox_oer"],
"workflow/feature_engineering",
"out_data")
file_name_i = "df_features_targets.pickle"
path_i = os.path.join(directory, file_name_i)
if not os.path.exists(directory): os.makedirs(directory)
with open(path_i, "wb") as fle:
pickle.dump(df_features_targets, fle)
# #########################################################
# +
from methods import get_df_features_targets
df_features_targets_tmp = get_df_features_targets()
df_features_targets_tmp.head()
# -
# #########################################################
print(20 * "# # ")
print("All done!")
print("Run time:", np.round((time.time() - ti) / 60, 3), "min")
print("combine_features_targets.ipynb")
print(20 * "# # ")
# #########################################################
# + active=""
#
#
#
def old_get_ORR_PLT():
"""
"""
#| - get_ORR_PLT
# #########################################################
# df_ads = get_df_ads()
# df_ads = df_ads[~df_ads.g_oh.isna()]
# df_m = df_ads
# #########################################################
df_dft = get_df_dft()
# #########################################################
df_job_ids = get_df_job_ids()
# #########################################################
df_features_targets = get_df_features_targets()
smart_format_dict = [
[{"stoich": "AB2"}, {"color2": "black"}],
[{"stoich": "AB3"}, {"color2": "grey"}],
]
ORR_PLT = ORR_Free_E_Plot(
free_energy_df=None,
state_title="ads",
free_e_title="ads_g",
smart_format=smart_format_dict,
color_list=None,
rxn_type="OER")
# # df_m.g_ooh = 1.16 * df_m.g_oh + 2.8
# df_m["g_ooh"] = df_m.g_oh + 2.8
# df_m = df_m.set_index(["compenv", "slab_id", ], drop=False)
new_col = (df_features_targets["targets"]["g_oh"] + 2.8)
new_col.name = ("targets", "g_ooh", "", )
df_features_targets = pd.concat([
new_col,
df_features_targets,
], axis=1)
paths_dict = dict()
# for name_i, row_i in df_m.iterrows():
for name_i, row_i in df_features_targets.iterrows():
#| - Loop through data and add to ORR_PLT
# #####################################################
g_o_i = row_i[("targets", "g_o", "", )]
g_oh_i = row_i[("targets", "g_oh", "", )]
g_ooh_i = row_i[("targets", "g_ooh", "", )]
slab_id_i = row_i[("data", "slab_id", "")]
active_site_i = row_i[("data", "active_site", "")]
job_id_o_i = row_i[("data", "job_id_o", "")]
job_id_oh_i = row_i[("data", "job_id_oh", "")]
# #####################################################
# #####################################################
df_job_ids_i = df_job_ids[df_job_ids.slab_id == slab_id_i]
bulk_ids = df_job_ids_i.bulk_id.unique()
mess_i = "SIJFIDSIFJIDSJIf"
assert len(bulk_ids) == 1, mess_i
bulk_id_i = bulk_ids[0]
# #########################################################
row_dft_i = df_dft.loc[bulk_id_i]
# #########################################################
stoich_i = row_dft_i.stoich
# #########################################################
data_dict_list = [
{"ads_g": g_o_i, "ads": "o", },
{"ads_g": g_oh_i, "ads": "oh", },
{"ads_g": g_ooh_i, "ads": "ooh", },
{"ads_g": 0., "ads": "bulk", },
]
df_i = pd.DataFrame(data_dict_list)
df_i["stoich"] = stoich_i
prop_name_list = [
"stoich",
]
# #########################################################
# name_i = "IDSJFISDf"
name_i = slab_id_i + "__" + str(int(active_site_i))
ORR_PLT.add_series(
df_i,
plot_mode="all",
overpotential_type="OER",
property_key_list=prop_name_list,
add_overpot=False,
name_i=name_i,
)
#__|
return(ORR_PLT)
