# lecture 5, seaborn basics

import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')

tips['total_cost'] = tips['total_bill'] + tips['tip']

# reset plotting image
plt.clf()

# create the actual plot, bins controls the number
# of categories (bars) in the plot!
sns.distplot(tips['total_cost'], bins=30)

# show the plot to user
plt.figure()


# draw second plot to its own image
plt.clf()

# if you don't want the kernel density estimation line,
# you can disable it!
sns.distplot(tips['tip'], kde=False)
plt.figure()

# first clear, then plot, then show figure
plt.clf()
sns.countplot(x='smoker', data=tips)
plt.figure()

# exact numbers of smokers and non-smokers
print(tips.groupby('smoker').count())

plt.clf()
sns.jointplot(x='total_bill', y='tip', data=tips)
plt.figure()

plt.clf()
sns.jointplot(x='total_bill', y='tip', data=tips, kind="reg")
plt.figure()


# NEW FILE

import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np


tips = sns.load_dataset('tips')
flights = sns.load_dataset('flights')

plt.clf()
sns.pairplot(tips)
plt.figure()

tips_correlations = tips.corr()

plt.clf()
sns.pairplot(flights)
plt.figure()

flights_correlations = flights.corr()

plt.clf()
# by default, barplot shows the average for selected category
sns.barplot(x='smoker', y='total_bill', data=tips, estimator=np.median)
plt.figure()

print(tips['smoker'].value_counts())
print(tips['day'].value_counts())
print(tips['time'].value_counts())

plt.clf()
sns.boxplot(x='time', y='total_bill', data=tips, hue='day')
plt.figure()

plt.clf()
sns.violinplot(x='day', y='tip', data=tips, hue='smoker', split=True)
plt.figure()

# NEW FILE

import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

tips = sns.load_dataset('tips')

plt.clf()

sns.stripplot(x='day', y='total_bill',
              data=tips, jitter=True, hue='smoker', split=True)

plt.figure()


plt.clf()
sns.swarmplot(x='day', y='total_bill', data=tips)
plt.figure()

plt.clf()

sns.violinplot(x='day', y='total_bill', data=tips)
sns.swarmplot(x='day', y='total_bill', data=tips, color='purple')

plt.figure()

# NEW FILE

import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

flights = sns.load_dataset('flights')
tips = sns.load_dataset('tips')

flights_correlations = flights.corr()
tips_correlations = tips.corr()


plt.clf()
sns.heatmap(tips_correlations, annot=True, cmap="magma")
plt.figure()


flights_pivot = flights.pivot_table(index='month', columns='year',
                                    values='passengers')

plt.clf()
sns.heatmap(flights_pivot, linecolor='black',
            linewidths=1)

plt.figure()

plt.clf()
sns.clustermap(flights_pivot, cmap='coolwarm', standard_scale=1)
plt.figure()

plt.clf()
sns.lmplot(x='total_bill', y='tip',
           data=tips, hue='smoker', row="sex",
           markers=['o', 'v'], scatter_kws={'s':100})
plt.figure()