import os.path
from sklearn.neighbors import KNeighborsClassifier
sys.path.insert(0, '.')
from optimize_model import optimize_model
submission_file = 'submission_knn_f3.csv'
features_files = [['data/Dog_1/features_02.txt']]
"""
features_files = [['data/Dog_1/features_02.txt',
'data/Dog_2/features_01.txt',
'data/Dog_3/features_01.txt',
'data/Dog_4/features_01.txt',
'data/Dog_5/features_01.txt',
'data/Patient_1/features_01.txt',
'data/Patient_2/features_01.txt']]
"""
for f in features_files:
if type(f) == str:
sys.exit('Each element of features_files must be a list of files.')
print f
optimize_model(f, submission_file, min_features=1, max_features=2,
classifier=KNeighborsClassifier,
parameters={'n_neighbors': [5, 10, 15],
'weights': ['uniform', 'distance']},
outlier_sigma=2, normalize_probs=None, n_cv=100)
raw_input('Press Enter to end.')
#!/usr/bin/env python
import sys
import os.path
sys.path.insert(0, '.')
from optimize_model import optimize_model
submission_file = 'submission_log_reg_f1to4_reopt_rocslope10.csv'
features_files = [['data/Dog_1/features_02.txt'],
['data/Dog_2/features_01.txt'],
['data/Dog_3/features_01.txt'],
['data/Dog_4/features_01.txt'],
['data/Dog_5/features_01.txt'],
['data/Patient_1/features_01.txt'],
['data/Patient_2/features_01.txt']]
for f in features_files:
if type(f) == str:
sys.exit('Each element of features_files must be a list of files.')
print f
optimize_model(f, submission_file, max_features=4,
outlier_sigma=2, normalize_probs='ROCSlope', n_cv=100)
raw_input('Press Enter to end.')
import os.path
from sklearn.svm import SVC
sys.path.insert(0, '.')
from optimize_model import optimize_model
submission_file = 'submission_svm_f1to2.csv'
features_files = [['data/Dog_1/features_02.txt'],
['data/Dog_2/features_01.txt'],
['data/Dog_3/features_01.txt'],
['data/Dog_4/features_01.txt'],
['data/Dog_5/features_01.txt'],
['data/Patient_1/features_01.txt'],
['data/Patient_2/features_01.txt']]
for f in features_files:
if type(f) == str:
sys.exit('Each element of features_files must be a list of files.')
print f
optimize_model(f, submission_file, min_features=1, max_features=2,
feature_columns=range(2, 27),
classifier=SVC,
parameters={'C': [0.1, 1, 10],
'gamma': [0.25, 0.5, 1],
'kernel': ['rbf'], 'probability': [True],
'class_weight': ['auto']},
outlier_sigma=2, normalize_probs='IsoReg', n_cv=20)
raw_input('Press Enter to end.')
import sys
import os.path
from sklearn.ensemble import RandomForestClassifier
sys.path.insert(0, '.')
from optimize_model import optimize_model
submission_file = 'submission_rf_f4to7imp_all.csv'
features_files = [['data/Dog_1/features_02.txt',
'data/Dog_2/features_01.txt',
'data/Dog_3/features_01.txt',
'data/Dog_4/features_01.txt',
'data/Dog_5/features_01.txt',
'data/Patient_1/features_01.txt',
'data/Patient_2/features_01.txt']]
for f in features_files:
if type(f) == str:
sys.exit('Each element of features_files must be a list of files.')
print f
optimize_model(f, submission_file, min_features=4, max_features=7,
feature_columns=[11, 5, 4, 10, 2, 9, 6],
classifier=RandomForestClassifier,
parameters={'n_estimators': [10, 50],
'criterion': ['entropy'],
'min_samples_leaf': [3, 6, 10]},
outlier_sigma=2, normalize_probs=None, n_cv=100)
raw_input('Press Enter to end.')
