PranjalDureja0002

s1 = data['score']
s_s1 = sum(s1.tolist())
s_s1

plt.style.use('fivethirtyeight')
ax=data.head(20).plot(kind = 'barh' , color = 'red')
for p in ax.patches:
        percentage = '{:.1f}%'.format(100 * p.get_width()/s_s1) 
        x = p.get_x() + p.get_width() - 0.5
        y = p.get_y() + p.get_height()

PranjalDureja0002

+---------+-----------------+----------------+----------------+---------------+
|  Model  | Train AUC Score | Test AUC Score | Train F1 Score | Test F1 Score |
+---------+-----------------+----------------+----------------+---------------+
| XgBoost |     0.99938     |    0.99855     |     0.9998     |    0.990791   |
+---------+-----------------+----------------+----------------+---------------+

PranjalDureja0002

x_cfl=XGBClassifier(n_estimators=1000,nthread=-1)
x_1=XGBClassifier(n_estimators=500,nthread=-1)
x_2=XGBClassifier(n_estimators=500,nthread=-1)

x_3 = DecisionTreeClassifier(max_depth=best_depth,min_samples_split=best_samples,class_weight='balanced')
x_4 = LogisticRegression(class_weight='balanced')
s_clf = StackingClassifier(classifiers=[x_1,x_2,x_3,x_4],meta_classifier=x_cfl)
s_clf.fit(X_train,y_train)
#sig_clf = CalibratedClassifierCV(x_cfl, method="sigmoid")
#sig_clf.fit(X_train_f, y_train_f)

PranjalDureja0002

+---------------------+-----------------+----------------+----------------+---------------+
|        Model        | Train AUC Score | Test AUC Score | Train F1 Score | Test F1 Score |
+---------------------+-----------------+----------------+----------------+---------------+
| Stacking_Classifier |     0.99537     |     0.9902     |    0.99429     |    0.98759    |
+---------------------+-----------------+----------------+----------------+---------------+

PranjalDureja0002

#In the 80% train set, split the train set into d1 and d2.(50-50).
d1,d2,y1,y2 = train_test_split(X_train,y_train,stratify=y_train,test_size=0.5,random_state=15)
d1 = d1.reset_index(drop=True)
d2 = d2.reset_index(drop=True)
y1 = y1.reset_index(drop=True)
y2 = y2.reset_index(drop=True)

def generating_samples(d1, y1):
    """From this d1,sampling with replacement is done
    """

PranjalDureja0002

+--------------------------------+-------------+---------------+
| Custom_Stacking_Implementation | Base Models | Test F1 Score |
+--------------------------------+-------------+---------------+
|                                |      50     |    0.980001   |
|                                |     100     |    0.981479   |
|                                |     150     |    0.982725   |
+--------------------------------+-------------+---------------+

PranjalDureja0002

#In the 80% train set, split the train set into d1 and d2.(50-50).
d1,d2,y1,y2 = train_test_split(X_train,y_train,stratify=y_train,test_size=0.5,random_state=15)
d1 = d1.reset_index(drop=True)
d2 = d2.reset_index(drop=True)
y1 = y1.reset_index(drop=True)
y2 = y2.reset_index(drop=True)

def generating_samples(d1, y1):
    """From this d1,sampling with replacement is done
    """

PranjalDureja0002

+---------+-----------------+----------------+----------------+---------------+
|  Model  | Train AUC Score | Test AUC Score | Train F1 Score | Test F1 Score |
+---------+-----------------+----------------+----------------+---------------+
| XgBoost |     0.99938     |    0.99855     |     0.9998     |    0.990791   |
+---------+-----------------+----------------+----------------+---------------+ +---------------+-----------------+----------------+----------------+---------------+
|     Model     | Train AUC Score | Test AUC Score | Train F1 Score | Test F1 Score |
+---------------+-----------------+----------------+----------------+---------------+
| Decision_Tree |      0.9967     |     0.9909     |    0.99314     |     0.9771    |
+---------------+-----------------+----------------+----------------+---------------+ +---------------------+-----------------+----------------+----------------+---------------+
|        Model        | Train AUC Score | Test AUC Score | Train F1 Score | Test F1 Score |

PranjalDureja0002

def find_best_threshold(threshold, fpr, tpr):
    t = threshold[np.argmax(tpr*(1-fpr))]     
    print("the maximum value of tpr*(1-fpr)", max(tpr*(1-fpr)), "for threshold", np.round(t,3))
    return t

def predict_with_best_t(proba, threshold):
    predictions = []
    for i in proba:                              
        if i>=threshold:
            predictions.append(1)

PranjalDureja0002

def final_fun_1(X):
    """ function takes raw data as input,preprocessing is done,
    feature engineering is performed and predictions made on the
    best model already trained"""
            
    d_beneficiary = pd.read_csv('health_cs_data/' + X[0])
    d_inpatient = pd.read_csv('health_cs_data/' + X[1])
    d_outpatient = pd.read_csv('health_cs_data/' + X[2])
    d_labels = pd.read_csv('health_cs_data/' + X[3])