Curso de Data Science

import pandas as pd data = pd.read_csv('data/cars.csv') data

print(data.dtypes)

manufacturer_name     object
model_name            object
transmission          object
color                 object
odometer_value         int64
year_produced          int64
engine_fuel           object
engine_has_gas          bool
engine_type           object
engine_capacity      float64
body_type             object
has_warranty            bool
state                 object
drivetrain            object
price_usd            float64
is_exchangeable         bool
location_region       object
number_of_photos       int64
up_counter             int64
feature_0               bool
feature_1               bool
feature_2               bool
feature_3               bool
feature_4               bool
feature_5               bool
feature_6               bool
feature_7               bool
feature_8               bool
feature_9               bool
duration_listed        int64
dtype: object

data.describe()

# Media del precio en USD price_mean = data['price_usd'].mean() print(f'Media del precio USD {price_mean}')

Media del precio USD 6639.971021255613

# Mediana del precio en USD price_median = data['price_usd'].median() print(f'Mediana del precio USD {price_median}')

Mediana del precio USD 4800.0

import matplotlib.pyplot as plt plt.rcParamsDefault['figure.figsize']

# Modificar estilos para los gráficos scale_x = 2 scale_y = 1.5 plt.rcParams['figure.figsize'] = (6.4*scale_x, 4.8*scale_y) plt.style.use('seaborn-darkgrid')

# Mostrar histograma de precios con 20 bins data['price_usd'].plot.hist(bins=20) plt.title('Histograma de precio USD') plt.xlabel('Precio en USD') plt.ylabel('Frecuencia') plt.show()

import seaborn as sns _, ax = plt.subplots() sns.histplot(ax=ax, data=data, x='price_usd', hue='manufacturer_name') plt.show()

/shared-libs/python3.7/py-core/lib/python3.7/site-packages/IPython/core/pylabtools.py:137: UserWarning: Creating legend with loc="best" can be slow with large amounts of data.
  fig.canvas.print_figure(bytes_io, **kw)

# Mostrar histograma por tipo de motor _, ax = plt.subplots() sns.histplot(ax=ax, data=data, x='price_usd', hue='engine_type') plt.show()

_, ax = plt.subplots() sns.histplot(ax=ax, data=data, x='price_usd', hue='engine_type', multiple='stack') plt.show()

# Agrupar por el tipo de motor y contar los datos data.groupby('engine_type').count()

Q7_data = data[(data['manufacturer_name'] == 'Audi') & (data['model_name'] == 'Q7')] Q7_data

_, ax = plt.subplots() sns.histplot(ax=ax, data=Q7_data, x='price_usd', hue='year_produced') plt.show()