How do I plot my data using matplotlib in Python?

Question

I've gotten the publication years of different books using this:

# -*- coding: utf-8 -*-
"""
Created on Fri Mar 22 13:12:11 2019

@author: Oppilas
"""
from __future__ import division
from matplotlib import pyplot as plt
from collections import Counter

import pandas as pd
import numpy as np
import re
import math


file = "BL-Flickr-Images-Book.csv"
df = pd.read_csv(file)
cnt = 0

for row in df['Date of Publication']:
    try:
        row += 0
    except TypeError:
        try:
            new_value = int(row)
            df.loc[cnt,'Date of Publication'] = new_value
        except ValueError:
            new_row = re.sub("\D","",row)
            df.loc[cnt,'Date of Publication'] = int(new_row[:4])
    cnt += 1


pub_years = []

for year in df['Date of Publication']:
    if math.isnan(year):
        continue
    else:
        if len(str(year)) >= 4:
            pub_years.append(year)

So, how do I plot this data sensibly using matplotlib? I've tried pyplot, but the graph line was all over the place. I also tried to look at the documentation for hist, but couldn't get it working.

Is the data I've extracted poor, or is it my lack of skill with matplotlib?

Answer 1

In general, you almost never need to iterate over rows to process your dataframe. You can just work on the columns directly. For example, this should work:

df.groupby('Year').count().plot(marker='o')

If you have some mangled dates, eg numbers like 61, 62, 63, etc instead of 1961, 1962, 1963, etc, then perhaps you can filter them out:

df.loc[df['Year']<100, 'Year'] = df['Year'] + 1900

I changed your column df['Date of publication'] to df['Year'] to make the examples a bit easier to read.