def plt_sheriffs_counties_month_heatmap(
save=False,
file_name='plt_sheriffs_counties_month_heatmap.png',
file_format='png'):
# just get the sheriffs
df = df_sheriffs()
# months
months = list(set(df.month_year.to_list()))
months.sort()
# counties
counties = list(set(df[common.SOURCE_COL].to_list()))
counties.sort()
matrix = pd.DataFrame(np.zeros(shape=(len(counties), len(months))),
columns=months,
index=counties)
for month, month_group in df.groupby(df.month_year):
for area, area_group in month_group.groupby(common.SOURCE_COL):
matrix.at[area, month] = 1
cmap = mpl.colors.ListedColormap(['#ffffff', '#0099ff'])
norm = mpl.colors.BoundaryNorm([0, 1, 3], cmap.N)
sns.heatmap(matrix, cmap=cmap, norm=norm, cbar=False)
# add labels etc.
set_labels_title('Month', 'Shires', 'Months that sheriffs appeared')
save_or_show(save, file_name, file_format)
def plot_radar(categories, values, title, colour, save=False, file_name='radar.png', file_format='png'):
""" Provide a radar chart. This borrows heavily from https://python-graph-gallery.com/390-basic-radar-chart/"""
n = len(categories)
values += values[:1]
angles = [x / float(n) * 2 * pi for x in range(n)]
angles += angles[:1]
# Initialise the spider plot
ax = plt.subplot(111, polar=True)
# Draw one axe per variable + add labels labels yet
plt.xticks(angles[:-1], categories, fontname="Times New Roman", color='black', size=10)
# Draw ylabels
ax.set_rlabel_position(0)
plt.yticks([10, 20, 30, 40], ["10", "20", "30", "40"], fontname="Times New Roman", color="black",
size=ANNOTATION_FONT_SIZE)
plt.ylim(0, 50)
# Plot data
ax.plot(angles, values, linewidth=2, linestyle='solid')
# Fill area
ax.fill(angles, values, colour, alpha=0.1)
plt.title(title, fontname="Times New Roman", fontsize=TITLE_FONT_SIZE, fontweight='bold')
save_or_show(save, file_name, file_format)
def plt_sheriffs_counties_terms_no_arrears_heatmap(
save=False,
file_name='plt_sheriffs_counties_terms_no_arrears_heatmap.png',
file_format='png'):
# just get the sheriffs
df = df_sheriffs()
df = df[~df[common.DETAILS_COL].str.contains('arrears')]
# counties
counties = list(set(df[common.SOURCE_COL].to_list()))
counties.sort()
matrix = pd.DataFrame(np.zeros(shape=(len(counties), len(common.TERMS))),
columns=common.TERMS,
index=counties)
for term, term_group in df.groupby(common.TERM_COL):
for area, area_group in term_group.groupby(common.SOURCE_COL):
matrix.at[area, term] = 1
cmap = mpl.colors.ListedColormap(['#ffffff', '#c28cb8'])
norm = mpl.colors.BoundaryNorm([0, 1, 3], cmap.N)
sns.heatmap(matrix, cmap=cmap, norm=norm, cbar=False)
# add labels etc.
set_labels_title('Terms', 'Shires',
'Terms that sheriffs appeared (arrears removed)')
save_or_show(save, file_name, file_format)
def plt_trans_count_week_term_pc_term_total(save=True, title='Number of weekly transactions as a % of the term total',
x_label="Week", y_label="Value of receipts as % of term total",
is_long_title=False, file='receipts_weekly_term_pc.png',
fig_size=PLOT_DIMENSIONS):
""" Create a line plot for the total number of transactions as a percentage of the total number of transactions
over the term. Each term is plotted on its own line. """
# set the style
sns.set(style=SN_STYLE, font=FONT_NAME)
# get data and remove 'NOTHING' values
df = roll.roll_with_entities_df()
df = filter_out_nothing(df)
plt.figure(figsize=fig_size)
for term, term_group in df.groupby(common.TERM_COL, sort=False):
# total for the term
total = term_group[common.PENCE_COL].count()
values = term_group.groupby(common.WEEK_COL)[common.PENCE_COL].count() / total * 100
sns.lineplot(y=values, x=values.index, label=term, marker='o')
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_pc_by_terms(save=True, title='% of the total payments, by term', x_label="Terms", y_label='% of total payments',
is_long_title=True, file='terms_total_pc.png', fig_size=PLOT_DIMENSIONS):
""" A basic bar graph that shows the payments per term as a percentage of the whole year. """
# set a plot size
plt.figure(figsize=fig_size)
# get the dataset
terms_df = roll.terms_overview_df()
# calculate the total income
total = terms_df['Term total'].sum()
# work out the %
term_total_as_pc = terms_df['Term total'] / total * 100
# plot the data
ax = sns.barplot(x=term_total_as_pc.index, y=term_total_as_pc)
# add the £.s.d. to each bar
for patch, pc in zip(ax.patches, term_total_as_pc):
ax.text(patch.get_x() + patch.get_width() / 2, patch.get_height(), '{0:.1f}%'.format(pc), ha="center",
fontname=FONT_NAME, fontsize=ANNOTATION_FONT_SIZE, linespacing=2.0)
plt.xticks(fontname=FONT_NAME)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_payments_days_swarm(save=True, title='Payments per day', x_label="Day of the week", y_label="Payment",
is_long_title=False, file='payments_day_swarm.png', fig_size=PLOT_DIMENSIONS, log=True):
""" Swarm plot showing the distribution of payments for each day. Each term has a different color."""
df = roll.roll_with_entities_df()
# set a plot size
plt.figure(figsize=fig_size)
ax = sns.stripplot(x=common.DAY_COL, y=common.PENCE_COL, hue=common.TERM_COL, data=df, jitter=0.3,
order=common.DAYS_OF_WEEK)
if log:
ax.set_ylim(bottom=1)
ax.set_yscale('log')
plt.legend(loc='lower right', shadow=True, fontsize='small')
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_business_by_day(save=True, title='Total no. of items of business occurring on each day',
x_label="Days of the week", y_label="No. of items of business", is_long_title=False,
file='business_by_day.png', fig_size=PLOT_DIMENSIONS, filter_nothing=True):
""" A plot that shows the total number of business entered on a day. """
df = roll.roll_with_entities_df()
# filter out the rows that have no receipts
if filter_nothing:
df = filter_out_nothing(df)
plt.figure(figsize=fig_size)
# set the style
sns.set(style=SN_STYLE, font=FONT_NAME)
# create the plot
sns.countplot(x=common.DAY_COL, data=df, order=common.DAYS_OF_WEEK)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_total_payments_by_source(save=True, title='Total payments by source',
x_label="Terms", y_label="Source", is_long_title=False,
file='total_payments_source.png', fig_size=PLOT_DIMENSIONS):
""" Bar plots showing the total amount received from each 'Source'. """
# set the style
sns.set(style="darkgrid")
plt.figure(figsize=fig_size)
df = roll.payments_overview_df()
sns.barplot(data=df, x=common.SOURCE_COL, y=common.PENCE_COL)
# tick range in pence (divisible by 240, which is £1)
ticks_range = np.arange(0, 480000, 120000)
# show the labels as £ rather than pennies
ticks_labels = []
for x in np.nditer(ticks_range.T):
ticks_labels.append(money.pence_to_psd(x))
plt.yticks(ticks_range, ticks_labels)
plt.xticks(fontsize=ANNOTATION_FONT_SIZE, fontname=FONT_NAME, rotation=90)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_scatter_payments_year(save=True, title='Scatter plot of all payments', x_label="Date",
y_label="Payment", is_long_title=False, file='scatter_plot_all_payments.png',
fig_size=PLOT_DIMENSIONS):
""" A scatter plot of payments across the year. """
# set a plot size
plt.figure(figsize=fig_size)
# get the data and remove rows with no payments
df = roll.roll_with_entities_df()
df['Date Time'] = df.apply(to_date, axis=1)
df_plot = df[df[common.PENCE_COL] > 0]
# create the plot
plt.scatter(df_plot['Date Time'].tolist(), df_plot['Pence'], s=2)
# tick range in pence (divisible by 240, which is £1)
ticks_range = np.arange(0, 84000, 6000)
# show the labels as £ rather than pennies
ticks_labels = []
for x in np.nditer(ticks_range.T):
ticks_labels.append(money.pence_to_psd(x))
plt.yticks(ticks_range, ticks_labels)
plt.xticks(fontsize=ANNOTATION_FONT_SIZE, rotation=90, fontname=FONT_NAME)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_business_by_term(save=True, title='Total number of items of business, per term', x_label="Terms",
y_label="No. of items of business", is_long_title=False, file='business_by_term.png',
fig_size=PLOT_DIMENSIONS):
""" A bar plot that groups by terms and shows number of items of business (profferings) for that term. """
# set a plot size
plt.figure(figsize=fig_size)
# get the totals
terms_df = roll.terms_overview_df()
# plot the data
ax = sns.barplot(x=terms_df.index, y=terms_df['Total entries'])
for patch, total in zip(ax.patches, terms_df['Total entries']):
ax.text(patch.get_x() + patch.get_width() / 2, patch.get_height(), total, ha="center",
fontname=FONT_NAME, fontsize=ANNOTATION_FONT_SIZE, linespacing=2.0)
plt.xticks(fontname=FONT_NAME)
plt.yticks(fontname=FONT_NAME)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_keyword_frequency(title='{} most frequent keywords',
x_label='Keyword',
y_label='Frequency',
limit=20,
save=False,
is_long_title=False,
file='plt_keyword_frequency.png',
fig_size=PLOT_DIMENSIONS):
"""" Display a plot that shows the most frequent keywords. The number of words is determined by the limit. """
# set the style
sns.set(style=SN_STYLE, font=FONT_NAME)
plt.figure(figsize=fig_size)
title = title.format(limit)
# get data
df = roll.roll_with_entities_df()
# filter out rows without keywords
df_keywords = df[df[common.KEYWORDS_COL].notnull()]
# get keywords as a list
all_keywords = df_keywords[common.KEYWORDS_COL].to_list()
# remove the semicolon delimiter
kw = []
for k in all_keywords:
for i in k.split(';'):
kw.append(i)
kw_freq = FreqDist(kw)
# create a pandas as df
kw_df = pd.DataFrame(kw_freq.most_common(limit),
columns=['Word', 'Frequency']).set_index('Word')
sns.set()
kw_df.plot(kind='bar', legend=None)
plt.xticks(fontsize=10, fontname=FONT_NAME)
plt.yticks(fontsize=10, fontname=FONT_NAME)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_total_by_terms(save=True, title='Total payments, per term', x_label="Terms", y_label='Total payments',
is_long_title=True, file='terms_total.png', fig_size=PLOT_DIMENSIONS):
""" A basic bar graph that shows the total payments per term. """
# set a plot size
plt.figure(figsize=fig_size)
# get the totals
terms_df = roll.terms_overview_df()
# tick range in pence (divisible by 240, which is £1)
ticks_range = np.arange(0, 600000, 120000)
# show the labels as £ rather than pennies
ticks_labels = []
for x in np.nditer(ticks_range.T):
ticks_labels.append(money.pence_to_psd(x))
plt.yticks(ticks_range, ticks_labels, fontname=FONT_NAME)
# plot the data
ax = sns.barplot(x=terms_df.index, y=terms_df['Term total'])
plt.xticks(fontname=FONT_NAME)
# add the £.s.d. to each bar
for patch, pence in zip(ax.patches, terms_df['Term total']):
ax.text(patch.get_x() + patch.get_width() / 2, patch.get_height(), money.pence_to_psd(pence), ha="center",
fontname=FONT_NAME, fontsize=ANNOTATION_FONT_SIZE, linespacing=2.0)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_source_term_pc_heat_map(save=True, title='Payments per term by source',
x_label="Total", y_label="Source", is_long_title=False,
file='total_payments_source_heatmap.png', fig_size=(10, 10)):
""" General heat map plot that shows the source of payments and the total values. """
df = roll.source_term_payments_pc_matrix_df()
# df_psd = df.applymap(money.pence_to_psd)
plt.figure(figsize=fig_size)
plt.rcParams['font.family'] = FONT_NAME
with sns.axes_style('white'):
hm = sns.heatmap(df, annot=True, cmap='Oranges', cbar=False, fmt='.1f', annot_kws={'size': ANNOTATION_FONT_SIZE})
hm.set_yticklabels(hm.get_yticklabels(), rotation=0)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
def plt_days_by_term(save=True, title='Number of days, per term', x_label="Terms", y_label="Days", is_long_title=False,
file='terms_days.png', fig_size=PLOT_DIMENSIONS):
""" A bar plot that groups by term with information about days. """
# set a plot size
plt.figure(figsize=fig_size)
# get the totals
terms_df = roll.terms_overview_df()
# plot the data
terms_df[['Total Days', 'Days with payments', 'Days with no payments']].plot(kind='bar', colormap='tab20b')
plt.xticks(fontname=FONT_NAME)
plt.yticks(fontname=FONT_NAME)
# plot labels
set_labels(x_label, y_label)
# plot title
title_text(title, is_long_title)
# show or save the image to file
save_or_show(save=save, plot_file_name=file)
