a_09_ejemplos

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Fri Feb 22 19:33:20 2019

@author: usrdel
"""


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

directorio = "./data/artwork_data.csv"

df_artwork = pd.read_csv(directorio,
                         nrows=5,
                         index_col="id",
                         usecols=["id","artist"])
# Archivos Texto PDF HTML XML CSV JSON
# Binarios
# BDD Relacionales

columnas = ["id", "artist", "title",
            "medium", "year", "acquisitionYear",
            "height", "width", "units"]
df_completo = pd.read_csv(
        directorio,
        index_col="id",
        usecols=columnas)

directorio_guardado = "./data/artwork_data.pickle"

df_completo.to_pickle(directorio_guardado)

df_completo_de_pickle = pd.read_pickle(
        directorio_guardado)

df_completo["Nueva Columna"] = 0

df_artistas = df_completo["artist"]


artistas = pd.unique(df_artistas) # DataFrame
len(df_completo)
len(artistas)


df_artistas_vc = df_artistas.value_counts()


df_completo["medium"].value_counts()

df_group_artista = df_completo.groupby("artist")

type(df_group_artista)

for name, group_fd in df_group_artista:
    print(name)
    print(group_fd)


df_completo['acquisitionYear'].mean() 


df_completo.tail(10)

df_completo.loc[1035,'artist']

df_completo.iloc[0:100,:]


df_completo.loc[:,'width'] = pd.to_numeric(
        df_completo['width'], errors = 'coerce')

df_completo.loc[:,'height'] = pd.to_numeric(
        df_completo['height'], errors = 'coerce')

df_completo["area"] = df_completo["width"] * df_completo["height"] 
                        


plt.plot(df_completo["height"] )

df_group_artista.agg(np.min)
df_group_artista.min()