jtemporal

import seaborn as sns

from sklearn.cluster import KMeans


# preparando nossos dados
iris = sns.load_dataset('iris')
df = iris.drop('species', axis=1)

# calculando a soma dos quadrados para as 19 quantidade de clusters

jtemporal

def optimal_number_of_clusters(wcss):
    x1, y1 = 2, wcss[0]
    x2, y2 = 20, wcss[len(wcss)-1]

    distances = []
    for i in range(len(wcss)):
        x0 = i+2
        y0 = wcss[i]
        numerator = abs((y2-y1)*x0 - (x2-x1)*y0 + x2*y1 - y2*x1)
        denominator = sqrt((y2 - y1)**2 + (x2 - x1)**2)

jtemporal

def calculate_wcss(data):
    wcss = []
    for n in range(2, 21):
        kmeans = KMeans(n_clusters=n)
        kmeans.fit(X=data)
        wcss.append(kmeans.inertia_)

    return wcss

jtemporal

import pandas as pd
import numpy as np

raw_data = {'id': [1,2,1,1,3,4], 'categoria': ['A', 'B', 'A', 'B', 'C', np.nan]}
t = pd.DataFrame(raw_data, columns=['id', 'categoria'])

def cat_mais_freq(user):
    value_count_index = user['categoria'].value_counts().index
    if value_count_index.empty:
        return None

jtemporal

#!/bin/bash

if [ -z ${DOCKER_USERNAME} ]; then
    echo "Missing DOCKER_USERNAME variable!"
    exit 1
fi

error() {
    if [ $? != 0 ]; then
        echo "Error!"

jtemporal

Esses dois codigos tem o mesmo resultado

paises <- c("Canadá","China","China","China","Estados Unidos",
          "Japão","Índia","Japão","Japão","México","Estados Unidos",
          "Índia","Estados Unidos","Estados Unidos","China","Alemanha",
          "Inglaterra","Estados Unidos","Japão","Inglaterra",
          "Estados Unidos","China","Índia","Canadá","Estados Unidos",
          "Índia","México","Índia","Suécia","Alemanha","China","China",
          "China","China","Estados Unidos","Japão","Japão","Alemanha",

jtemporal

class Cachorrinho(Model):
    nome = StringType(required=True, deserialize_from="name",
                      serialized_name="name")
    cor = StringType(required=True, deserialize_from="fur",
                      serialized_name="fur")
    def validate_nome(self, data, value):
        if data['nome'].capitalize() != data['nome']:
            raise ValidationError('Pet name is not capitalized!')
        return value

jtemporal

from schematics.exceptions import ValidationError
from schematics.models import Model
from schematics.types import StringType, ListType, ModelType, IntType

from udtypes import SizedNumberType


class Cachorrinho(Model):
    nome = StringType(required=True, deserialize_from="name",
                      serialized_name="name")

jtemporal

from schematics.models import Model
from schematics.types import StringType, ListType, ModelType, IntType

from udtypes import SizedNumberType


class Cachorrinho(Model):
    nome = StringType(required=True, deserialize_from="name",
                      serialized_name="name")
    cor = StringType(required=True, deserialize_from="fur",

jtemporal

from schematics.types import IntType
from schematics.exceptions import ValidationError

class SizedNumberType(IntType):
    def validate_number(self, number):
        sized = str(number)
        if len(sized) != 2:
            raise ValidationError("This number doesn't comply with the expected format size 2")
        return number