artists_data_cleaning_lab

# delete existing copies
#! rm *.csv ;

# Get the data from GitHub
!wget https://gist.githubusercontent.com/douglasgoodwin/4bebb2445759313a95690de41d338729/raw/5214a3761f648ccc9f3d692eab8a8e9cc8cd3fd7/artists.csv ;
!wget https://gist.githubusercontent.com/douglasgoodwin/4bebb2445759313a95690de41d338729/raw/5214a3761f648ccc9f3d692eab8a8e9cc8cd3fd7/artworks.csv

from collections import Counter
import random
from datetime import datetime, date
from dataclasses import dataclass, field
from typing import Optional, List, Dict

# Plotting cell
import seaborn as sns
from matplotlib import pyplot as plt

# font
plt.rcParams.update({'font.size': 8})

# reset the default figsize value
plt.rcParams["figure.figsize"] = plt.rcParamsDefault["figure.figsize"]

# 144 is good for a high-resolution display. Try 100 if it's too big
plt.rcParams["figure.dpi"] = (120)

# code to draw a donut plot
# Make a function to display a donut plot

def donutplot(keys=["yes","no"],vals=[33,20],legend="Are you in a good mood?"):
  # Create a white circle at the center of the plot
  my_circle = plt.Circle( (0,0), 0.7, color='white')

  # Draw pie wedges with thick white edges
  props = {'linewidth':4, 'edgecolor':'white'}
  plt.pie(vals,labels=keys,wedgeprops=props )

  plt.title(legend)
  p = plt.gcf()   # get current figure
  p.gca().add_artist(my_circle)

  return p

# Utility functions

import csv
from pathlib import Path

def loadData(source_path: Path) -> List[Dict[str, str]]:
  """Useful fuction to load local CSV files"""
  with source_path.open() as source_file:
    rdr= csv.DictReader(source_file)
    data: List[Dict[str, str]] = list(rdr)
    return data

def string2Date(astring='2000-01-19'):
  """The dates are a mess, clean them up! Did I miss anything?"""
  try:
    if astring == '':
      return None
    elif 'Unknown' in astring:
      return None
    elif 'c. ' in astring:
      # 'c. 1917'
      y = astring.split(' ')[-1]
      dto = datetime.strptime(y, '%Y')
      return dto
    elif astring == 'n.d.':
      return None
    elif len(astring.split('-')) == 3:
      dto = datetime.strptime(astring, '%Y-%m-%d')
      return dto
    elif len(astring.split('–')) == 2:
      # ie date is 1977–78 -- a long hyphen
      y = astring.split('–')[0]
      dto = datetime.strptime(y, '%Y')
      return dto
    elif len(astring.split('-')) == 2:
      # ie date is 1977-78
      y = astring.split('-')[0]
      dto = datetime.strptime(y, '%Y')
      return dto
    else:
      dto = datetime.strptime(astring, '%Y')
      return dto
  except:
    return None

def string2Float(astring):
  if astring == '':
    return None
  else:
    return float(astring)

@dataclass
class Artist:
    id: int
    name: str=None
    nationality: str=None
    gender: str=None
    birth: datetime=None
    death: datetime=None

    def getMyArtworks(self,artworks):
      myartworks = []
      for w in artworks:
        if self.id in w.artists:
          myartworks.append(w)
      return myartworks



@dataclass
class Artwork:
    id: int
    title: str="Untitled"
    # artist: List[Artist]=field(default_factory=list)
    artists: List[int]=field(default_factory=list)
    name: str="unspecified"
    date: datetime=None
    medium: str=None
    dimensions: str=None
    acquired: datetime=None
    credit: str=None
    catalogue: str=None
    department: str=None
    classification: str=None
    objectnumber: int=None
    diameter: int=None
    circumference: int=None
    height: int=None
    length: int=None
    width: int=None
    depth: int=None
    weight: int=None
    duration: int=None

    def artistDetail(self,artists):
      myartists = []
      for a in artists:
        if a.id in self.artists:
          myartists.append(a)
      return myartists

artists_csv = loadData(Path("artists.csv"))
artists = []

for a in artists_csv:
  if a['Nationality'] != '':
    nation = a['Nationality']
  else:
    nation = None
  artists.append(Artist(int(a['Artist ID']),
                        a['Name'],
                        nation,
                        a['Gender'],
                        string2Date(a['Birth Year']),
                        string2Date(a['Death Year'])))

artworks_csv = loadData(Path("artworks.csv"))
artworks = []

# Artwork ID,Title,Artist ID,Name,Date,Medium,Dimensions,Acquisition Date,Credit,
# Catalogue,Department,Classification,Object Number,Diameter (cm),Circumference (cm),Height (cm),
# Length (cm),Width (cm),Depth (cm),Weight (kg),Duration (s)

for w in artworks_csv:
  # artworks may have many artists
  artistids = w['Artist ID'].split(',')
  aids = []
  for aid in artistids:
    try:
      aids.append(int(aid))
    except:
      pass

  artworks.append(Artwork(int(w['Artwork ID']),
                  w['Title'],
                  aids,
                  w['Name'],
                  string2Date(w['Date']),
                  w['Medium'],
                  w['Dimensions'],
                  string2Date(w['Acquisition Date']),
                  w['Credit'],
                  w['Catalogue'],
                  w['Department'],
                  w['Classification'],
                  w['Object Number'],
                  string2Float(w['Diameter (cm)']),
                  string2Float(w['Circumference (cm)']),
                  string2Float(w['Height (cm)']),
                  string2Float(w['Length (cm)']),
                  string2Float(w['Width (cm)']),
                  string2Float(w['Depth (cm)']),
                  string2Float(w['Weight (kg)']),
                  string2Float(w['Duration (s)'])
                  )
  )

# sanity checks
# select a random artist
aid = random.randint(1,1000)
me = artists[aid]

myworks = me.getMyArtworks(artworks)
worktitles = "\n - ".join([w.title for w in myworks])

print(f"My name is {me.name}, my gender is {me.gender}, and I am {me.nationality}.")
print(f"MoMA collected {len(myworks)} of my works: \n - {worktitles}")

genderSpecified = []

for artist in artists:
  if artist.gender:
    genderSpecified.append(artist.gender)

c = Counter(genderSpecified)

d = donutplot(
    list(c.keys()),
    list(c.values()),
    legend=f"Gender of {len(genderSpecified):,} artists with specified gender")

plt.show()

noGender = [a.name for a in artists if a.gender == None]
noNationality = [a.id for a in artists if a.nationality == None]

print(f"{len(noGender):,} artists have unspecified gender.")
print(f"{len(noNationality):,} artists have unspecified nationality")
print()

# Start here
noName = []
noBirth = []
noDeath = []

for a in artists:
    if a.name == None:
        noName.append(a)
    if a.birth == None:
        noBirth.append(a)
    if a.death == None:
        noDeath.append(a)

ct = 0
for id in noName+noBirth+noDeath:
        ct += 1
        if ct < 100:
            print(id)
print()
        
print("Get these values:")
print(f"{len(noName):,} artists have no name.")
print(f"{len(noBirth):,} artists have no birth year.")
print(f"{len(noDeath):,} artists have no death year.")

# What are their nationalities?

c = Counter( [n.nationality for n in artists] )

nations = [n for n, counte in c.items() if counte >= 500]
counts = [counte for n, counte in c.items() if counte >= 500]

d = donutplot(
    list(nations),
    list(counts),
    legend=f"Top Nationalities of the {len(artists):,} artists collected by MoMA")


plt.show()

sns.barplot(x = nations, y = counts)

# Create a function that will print walltext for a randomly seelcted artwork
awid = random.randint(1,10000)
showme = artworks[awid]

# Display relevant info. Here's a hint!
print(f"{showme.title} | {showme.date:%Y}")
print(f"By {showme.name}")
print(f"Medium: {showme.medium}")
print(f"Dimensions: {showme.dimensions}")

print("Help, I need names!")

# How will you get the artists names?