JLFDataScience

#Import libraries
import pandas as pd
import numpy as np
import requests

#Test the Yandex API connection
key='APIKEY' #Enter the Apikey supplied to the registrar in Yandex
text = 'Hola Mundo'
lang = 'en'
url_yandex ="https://translate.yandex.net/api/v1.5/tr.json/translate?key=%s&text=%s&lang=%s" % (key,text,lang)

JLFDataScience

#Function to automate translation Yandex taslate
def traslate(text, key):
    lang = 'en'
    url_yandex ="https://translate.yandex.net/api/v1.5/tr.json/translate?key=%s&text=%s&lang=%s" % (key,text,lang)
    time.sleep(0.3)
    response = requests.get(url_yandex, timeout=None)
    response_data = eval(response.content.decode('utf-8'))
    lb = response_data['text'][0]
    return lb
 

JLFDataScience

dummie = pd.get_dummies(myopia['AGE'], prefix = 'AGE')
names = list(dummie.columns)
names.remove(names[0])
features.remove('AGE')
myopia_dummy = pd.concat([myopia[features], dummie[names]], axis = 1)

JLFDataScience

from sklearn.linear_model import LinearRegression

def calculateVIF(data):
    features = list(data.columns)
    num_features = len(features)
    
    model = LinearRegression()
    
    result = pd.DataFrame(index = ['VIF'], columns = features)
    result = result.fillna(0)

JLFDataScience

from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import f_classif

var_sk = SelectKBest(f_classif, k = 8)
var_sk.fit_transform(vif_selection, myopia[target])

k_selection = vif_selection.loc[:, var_sk.get_support()]
k_selection.head()

JLFDataScience

from sklearn.metrics import accuracy_score
from sklearn.metrics import auc
from sklearn.metrics import confusion_matrix
from sklearn.metrics import precision_score
from sklearn.metrics import recall_score
from sklearn.metrics import roc_curve

def metrics(y_true, y_pred):
    cm = confusion_matrix(y_true, y_pred)

JLFDataScience

from sklearn.model_selection import train_test_split

features = list(k_selection.columns)
x_train, x_test, y_train, y_test = train_test_split(k_selection, myopia[target],  random_state = 0)

JLFDataScience

from sklearn.linear_model.logistic import LogisticRegression

model = LogisticRegression().fit(x_train, y_train)
y_pred_train = model.predict(x_train)

metrics(y_train, y_pred_train)

JLFDataScience

from sklearn.tree import DecisionTreeClassifier

tree = DecisionTreeClassifier(criterion = 'entropy',
                              max_depth = 2,
                              random_state = 0)
tree.fit(x_train, y_train)
y_pred_train = tree.predict(x_train)

metrics(y_train, y_pred_train)

JLFDataScience

rf_classifier = RandomForestClassifier(n_estimators = 5, max_depth=4,
                                       random_state = 1)
rf_classifier.fit(x_train, y_train)
y_pred = rf_classifier.predict(x_train)

metrics(y_train, y_pred)