def c_index_multiple_from_python(matrix,
isdead,
nbdays,
matrix_test,
isdead_test,
nbdays_test,
isfactor=False):
"""
"""
frame = pd.DataFrame(matrix)
frame["isdead"] = isdead
frame["nbdays"] = nbdays
frame_test = pd.DataFrame(matrix_test)
frame_test["isdead"] = isdead_test
frame_test["nbdays"] = nbdays_test
cph = CoxPHFitter()
try:
with warnings.catch_warnings():
warnings.simplefilter("ignore")
cph.fit(frame, "nbdays", "isdead")
except Exception as e:
print(e)
return np.nan
cindex = cph.score(frame_test, scoring_method="concordance_index")
return cindex
def fit_cox(
train_df: Union[pd.DataFrame, str],
covariates: List[str],
test_df: Union[pd.DataFrame, str] = None,
strata: List[str] = None,
plot: bool = False,
process_dir: str = None,
):
if isinstance(train_df, str):
train_df = pd.read_csv(train_df)
if isinstance(test_df, str):
test_df = pd.read_csv(test_df)
cphf = CoxPHFitter()
included_cols = ["duration", "event"] + list(covariates)
print(train_df.columns)
cphf.fit(
train_df[included_cols],
duration_col="duration",
event_col="event",
strata=strata,
)
results = {
"log_likelihood":
cphf.log_likelihood_,
"concordance_index":
cphf.concordance_index_,
"log_likelihood_ratio_test_pvalue":
cphf.log_likelihood_ratio_test().p_value,
}
if test_df is not None:
results["test_log_likelihood"] = cphf.score(
test_df[included_cols], scoring_method="log_likelihood")
results["test_concordance_index"] = cphf.score(
test_df[included_cols], scoring_method="concordance_index")
if plot and process_dir is not None:
plt.figure(figsize=(5, 10))
cphf.plot()
plt.savefig(os.path.join(process_dir, "hazard_plot.pdf"))
if process_dir is not None:
cphf.summary.to_csv(os.path.join(process_dir, "summary.csv"))
save_dict_to_json(os.path.join(process_dir, "results.json"), results)
return results, cphf
def objective(trial, df_trn, df_val):
penalizer = trial.suggest_loguniform('penalizer', 1e-5, 1e2)
l1_ratio = trial.suggest_uniform('l1_ratio', 0.0, 1.0)
print(trial.params)
print('Fitting...')
cph = CoxPHFitter(penalizer=penalizer, l1_ratio=l1_ratio)
cph.fit(df_trn, duration_col='TIME', event_col='EVENT', show_progress=True)
print('done')
loglike = cph.score(df_val)
return loglike
def c_index_from_python(values,
isdead,
nbdays,
values_test,
isdead_test,
nbdays_test,
isfactor=False):
"""
"""
if isfactor:
values = np.asarray(values).astype("str")
values_test = np.asarray(values_test).astype("str")
frame = pd.DataFrame({
"values": values,
"isdead": isdead,
"nbdays": nbdays
})
frame_test = pd.DataFrame({
"values": values_test,
"isdead": isdead_test,
"nbdays": nbdays_test
})
cph = CoxPHFitter()
try:
with warnings.catch_warnings():
warnings.simplefilter("ignore")
cph.fit(frame, "nbdays", "isdead")
except Exception as e:
print(e)
return np.nan
cindex = cph.score(frame_test, scoring_method="concordance_index")
return cindex
# -*- coding: utf-8 -*-
# cox regression
if __name__ == "__main__":
import pandas as pd
import time
import numpy as np
from lifelines import CoxPHFitter
from lifelines.datasets import load_rossi, load_regression_dataset
reps = 1
df = load_rossi()
# df['s'] = "a"
df = pd.concat([df] * reps)
print(df.shape)
cph = CoxPHFitter(baseline_estimation_method="spline",
n_baseline_knots=3,
strata=["wexp"])
start_time = time.time()
cph.fit(df,
duration_col="week",
event_col="arrest",
show_progress=True,
timeline=np.linspace(1, 60, 100))
print(cph.score(df))
print("--- %s seconds ---" % (time.time() - start_time))
# special for Cox
xy_train_df = pd.DataFrame(x_train)
xy_train_df['T'] = y_train
xy_train_df['E'] = e_train
xy_val_df = pd.DataFrame(x_val)
xy_val_df['T'] = y_val
xy_val_df['E'] = e_val
xy_val_df_events = xy_val_df[xy_val_df['E'] == 1]
cph = CoxPHFitter(penalizer=0.1).fit(xy_train_df, 'T', 'E')
preds = -cph.predict_partial_hazard(xy_val_df)
cindex_train = cph.score(xy_train_df, scoring_method='concordance_index')
cindex_val = cph.score(xy_val_df, scoring_method='concordance_index')
cindex_val_events = cph.score(xy_val_df_events,
scoring_method='concordance_index')
cph_cindex_trains.append(cindex_train)
cph_cindex_vals.append(cindex_val)
cph_cindex_vals_events.append(cindex_val_events)
ps.append(p_to_drop)
print('Train cindex {:.2f}'.format(cindex_train * 100))
print('Test cindex {:.2f}'.format(cindex_val * 100))
print('Test cindex Events Only {:.2f}'.format(cindex_val_events * 100))
print(
'=================================================================================================='
)
# -*- coding: utf-8 -*-
# cox regression
if __name__ == "__main__":
import pandas as pd
import time
import numpy as np
from lifelines import CoxPHFitter
from lifelines.datasets import load_rossi, load_regression_dataset
reps = 1
df = load_rossi()
df = pd.concat([df] * reps)
cp_breslow = CoxPHFitter(penalizer=0.1,
l1_ratio=1.0,
baseline_estimation_method="spline")
start_time = time.time()
cp_breslow.fit(df,
duration_col="week",
event_col="arrest",
show_progress=True)
print("--- %s seconds ---" % (time.time() - start_time))
cp_breslow.print_summary(2)
print(cp_breslow.score(df))
print(cp_breslow.score(df, scoring_method="concordance_index"))
