mardani72

X_blind = blind[['Depth', 'GR', 'ILD_log10','DeltaPHI', 'PHIND', 'PE', 'NM_M', 'RELPOS', 'Formation_num']]
y_blind = blind['Facies']
# scale
scaler = StandardScaler()
X_blind = scaler.fit_transform(X_blind)


# define Classifiers
log = LogisticRegression(C = 10, solver = 'lbfgs', max_iter= 300 ) 
knn = KNeighborsClassifier(leaf_size = 10, n_neighbors=2)

mardani72

from itertools import cycle
from sklearn import svm, datasets
from sklearn.metrics import roc_curve, auc
from sklearn.preprocessing import label_binarize
from sklearn.multiclass import OneVsRestClassifier
from scipy import interp
from sklearn.metrics import roc_auc_score

# Binarize the output
yy = label_binarize(y_sm, classes=[ 1, 2, 3, 4, 5, 6, 7, 8, 9])

mardani72

from itertools import cycle
from sklearn import svm, datasets
from sklearn.metrics import roc_curve, auc
from sklearn.preprocessing import label_binarize
from sklearn.multiclass import OneVsRestClassifier
from scipy import interp
from sklearn.metrics import roc_auc_score
from sklearn.model_selection import learning_curve
from sklearn.model_selection import ShuffleSplit

mardani72

from sklearn.metrics import confusion_matrix
import itertools

# define function to implement confusion matrix with normalization capability
def plot_confusion_matrix(cm, classes, normalize=False,
                          title='Confusion matrix',
                          cmap=plt.cm.Reds):

    if normalize:
        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]

mardani72

from sklearn.metrics import classification_report
from sklearn.model_selection import KFold
from sklearn.model_selection import cross_validate
from sklearn.metrics import precision_recall_fscore_support

# define Classifiers with optimum hyper_param
log = LogisticRegression(C = 10, solver = 'lbfgs', max_iter= 200 ) 
knn = KNeighborsClassifier(leaf_size = 10, n_neighbors=2)
dtree = DecisionTreeClassifier(criterion = 'entropy', max_depth=15)
rtree = RandomForestClassifier(criterion='entropy', max_depth=20, n_estimators=300)

mardani72

from sklearn.model_selection import GridSearchCV
X_train, X_test, y_train, y_test = train_test_split(X_sm, y_sm, test_size=0.2, random_state=42)

#----------------------------------------------------------------logistic regression classifier
#define hyper parameters and ranges
param_grid_log = [{'C': [0.1, 1, 10], 'solver': ['lbfgs', 'liblinear'], 
                   'max_iter':[100, 300]}]
#apply gridsearch
grid_log  = GridSearchCV(log, param_grid=param_grid_log, cv=5)
#fit model with grid search

mardani72

from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.svm import SVC
from sklearn.naive_bayes import GaussianNB
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.ensemble import ExtraTreesClassifier

# define Classifiers

mardani72

from sklearn.inspection import permutation_importance
model = LogisticRegression(solver='liblinear')
# fit the model
model.fit(X, y)
# perform permutation importance
results = permutation_importance(model, X, y, scoring='accuracy')
# get importance
importance = results.importances_mean
# summarize feature importance
for i,v in enumerate(importance):

mardani72

from sklearn.tree import DecisionTreeClassifier
model = DecisionTreeClassifier()
# fit the model
model.fit(X, y)
# get importance
importance = model.feature_importances_
# summarize feature importance
for i,v in enumerate(importance):
	print('Feature: %0d, Score: %.5f' % (i,v))
# plot feature importance

mardani72

# logistic regression for feature importance
from sklearn.datasets import make_classification
from sklearn.linear_model import LinearRegression
from matplotlib import pyplot
# define dataset
model = LinearRegression()
# fit the model
model.fit(X, y)
# get importance
importance = model.coef_