school = 'Platzi' #print(____) # Incluir la variable declarada anteriormente.

import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns # agrega librerías que necesites

url_wine_red = 'https://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/winequality-red.csv' url_wine_white = 'https://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/winequality-white.csv'

red = pd.read_csv(url_wine_red, delimiter=";") white = pd.read_csv(url_wine_white, delimiter=";")

red.head(5)

white.head(5)

red['category']='red' white['category']='white'

red.shape

white.shape

total_wine=red.append(white, ignore_index=True)

total_wine.shape

total_wine.head(5)

total_wine.shape

total_wine.dtypes

total_wine.isnull().sum()

total_wine.describe()

quality = total_wine['quality'] quality.describe()

quality.value_counts()

countplot_quality = sns.countplot(quality) countplot_quality.set_title('Histograma de quality')

def calidad(x): #if x == 3 or x == 4: if x == 3 or x == 4 or x == 5: return ('Poor') #elif x == 5 or x == 6 or x == 7: elif x == 6: return ('Medium') else: return ('High') total_wine['quality_category'] = total_wine['quality'].apply(lambda x: calidad(x))

total_wine.quality_category = total_wine.quality_category.astype('category') total_wine.info()

countplot_quality_category=sns.countplot(total_wine['quality_category'],order=['Poor','Medium','High']) countplot_quality_category.set_title('Histograma de quality category')

for i in range(11): sns.set(rc={'figure.figsize':(11.7,8.27)}) f, (ax_hist,ax_boxt, ax_box) = plt.subplots(3, sharex=True, gridspec_kw={"height_ratios": (.5, .5,.5)}) sns.histplot(total_wine.iloc[:,i], ax=ax_hist) sns.boxplot(total_wine.iloc[:,i],ax=ax_boxt) sns.boxplot(x=total_wine.iloc[:,i], y='quality_category',data=total_wine,ax=ax_box) ax_hist.set_title(total_wine.iloc[:,i].name, size=22) ax_boxt.set(xlabel='')

total_wine_poor_or = total_wine[total_wine.quality_category=='Poor'] total_wine_poor_or = total_wine_poor_or.reset_index(drop=True) total_wine_poor_or['New_ID'] = total_wine_poor_or.index + 1

total_wine_poor = total_wine[total_wine.quality_category=='Poor'] total_wine_poor = total_wine_poor.reset_index(drop=True) total_wine_poor['New_ID'] = total_wine_poor.index + 1 t_w_c= [] for i in range (11): Q1 = total_wine_poor.iloc[:,i].quantile(q=0.25) Q3 = total_wine_poor.iloc[:,i].quantile(q=0.75) iqr = Q3 - Q1 minlimit = Q1 - 1.5*iqr maxlimit = Q3 + 1.5*iqr total_wine_filteredpoor = total_wine_poor[(total_wine_poor.iloc[:,i] < maxlimit) & (total_wine_poor.iloc[:,i] > minlimit)] t_w_c.append(total_wine_filteredpoor['New_ID']) #total_wine_poor=total_wine_filteredpoor

identidad=(pd.DataFrame(t_w_c[0:11])).T identidad1=identidad.dropna() identidadf=identidad1.iloc[:,0]

total_wine_poor_wo = total_wine_poor_or[total_wine_poor_or.New_ID.isin(identidadf)]

total_wine_medium_or = total_wine[total_wine.quality_category=='Medium'] total_wine_medium_or = total_wine_medium_or.reset_index(drop=True) total_wine_medium_or['New_ID'] = total_wine_medium_or.index + 1

total_wine_medium = total_wine[total_wine.quality_category=='Medium'] total_wine_medium = total_wine_medium.reset_index(drop=True) total_wine_medium['New_ID'] = total_wine_medium.index + 1 t_w_c_m= [] for i in range (11): Q1 = total_wine_medium.iloc[:,i].quantile(q=0.25) Q3 = total_wine_medium.iloc[:,i].quantile(q=0.75) iqr = Q3 - Q1 minlimit = Q1 - 1.5*iqr maxlimit = Q3 + 1.5*iqr total_wine_filteredmedium = total_wine_medium[(total_wine_medium.iloc[:,i] < maxlimit) & (total_wine_medium.iloc[:,i] > minlimit)] t_w_c_m.append(total_wine_filteredmedium['New_ID'])

identidadm=(pd.DataFrame(t_w_c_m[0:11])).T identidad1m=identidadm.dropna() identidadfm=identidad1m.iloc[:,0]

total_wine_medium_wo = total_wine_medium_or[total_wine_medium_or.New_ID.isin(identidadfm)]

total_wine_high_or = total_wine[total_wine.quality_category=='High'] total_wine_high_or = total_wine_high_or.reset_index(drop=True) total_wine_high_or['New_ID'] = total_wine_high_or.index + 1

total_wine_high = total_wine[total_wine.quality_category=='High'] total_wine_high = total_wine_high.reset_index(drop=True) total_wine_high['New_ID'] = total_wine_high.index + 1 t_w_c_h= [] for i in range (11): Q1 = total_wine_high.iloc[:,i].quantile(q=0.25) Q3 = total_wine_high.iloc[:,i].quantile(q=0.75) iqr = Q3 - Q1 minlimit = Q1 - 1.5*iqr maxlimit = Q3 + 1.5*iqr total_wine_filteredhigh = total_wine_high[(total_wine_high.iloc[:,i] < maxlimit) & (total_wine_high.iloc[:,i] > minlimit)] t_w_c_h.append(total_wine_filteredhigh['New_ID'])

identidadh=(pd.DataFrame(t_w_c_h[0:11])).T identidad1h=identidadh.dropna() identidadfh=identidad1h.iloc[:,0]

total_wine_high_wo = total_wine_high_or[total_wine_high_or.New_ID.isin(identidadfh)]

total_wine_final=total_wine_poor_wo.append(total_wine_medium_wo).append(total_wine_high_wo) total_wine_final=total_wine_final.reset_index(drop=True)

for i in range(11): sns.set(rc={'figure.figsize':(11.7,8.27)}) f, (ax_hist,ax_boxt, ax_box,ax_boxo) = plt.subplots(4, sharex=True, gridspec_kw={"height_ratios": (.5,.5, .5,.5)}) sns.histplot(total_wine_final.iloc[:,i], ax=ax_hist) sns.boxplot(total_wine_final.iloc[:,i],ax=ax_boxt) sns.boxplot(x=total_wine_final.iloc[:,i], y='quality_category',data=total_wine_final,ax=ax_box) sns.boxplot(x=total_wine.iloc[:,i], y='quality_category',data=total_wine,ax=ax_boxo) ax_boxt.set(xlabel='') ax_box.set(xlabel='') ax_hist.set_title(total_wine.iloc[:,i].name, size=22)

correlation_matriz=total_wine_final.drop(['New_ID'],axis=1).corr() correlation_matriz

sns.heatmap(correlation_matriz)

correlation_matriz['quality'].sort_values(ascending=False)

correlated_quality=total_wine_final[['alcohol','density','chlorides','volatile acidity','quality_category','category']]

correlation_matriz_quality=correlated_quality.corr() correlation_matriz_quality

sns.heatmap(correlation_matriz_quality)

sns.pairplot(correlated_quality,hue='quality_category')

sns.set(rc={'figure.figsize':(4,4)}) f, (ax_bar,ax_vio) = plt.subplots(2, sharex=True, gridspec_kw={"height_ratios": (.6,.6)}) sns.barplot(x='quality_category', y='alcohol', data=total_wine_final, ax=ax_bar,order=['Poor','Medium','High']) sns.violinplot(x='quality_category', y='alcohol',data=total_wine_final,ax=ax_vio,order=['Poor','Medium','High']) ax_bar.set(xlabel='') ax_bar.set_ylim(7,13)

sns.set(rc={'figure.figsize':(4,4)}) f, (ax_bar,ax_vio) = plt.subplots(2, sharex=True, gridspec_kw={"height_ratios": (.6,.6)}) sns.barplot(x='quality_category', y='density', data=total_wine_final, ax=ax_bar,order=['Poor','Medium','High']) sns.violinplot(x='quality_category', y='density',data=total_wine_final,ax=ax_vio,order=['Poor','Medium','High']) ax_bar.set(xlabel='') ax_bar.set_ylim(0.990,0.999)

sns.set(rc={'figure.figsize':(4,4)}) f, (ax_bar,ax_vio) = plt.subplots(2, sharex=True, gridspec_kw={"height_ratios": (.6,.6)}) sns.barplot(x='quality_category', y='chlorides', data=total_wine_final, ax=ax_bar,order=['Poor','Medium','High']) sns.violinplot(x='quality_category', y='chlorides',data=total_wine_final,ax=ax_vio,order=['Poor','Medium','High']) ax_bar.set(xlabel='') ax_bar.set_ylim(0.02,0.07)

sns.set(rc={'figure.figsize':(4,4)}) f, (ax_bar,ax_vio) = plt.subplots(2, sharex=True, gridspec_kw={"height_ratios": (.6,.6)}) sns.barplot(x='quality_category', y='volatile acidity', data=total_wine_final, ax=ax_bar,order=['Poor','Medium','High']) sns.violinplot(x='quality_category', y='volatile acidity',data=total_wine_final,ax=ax_vio,order=['Poor','Medium','High']) ax_bar.set(xlabel='') ax_bar.set_ylim(0.1,0.5)