test_and_case_data.py

   1 #!/usr/bin/env python
   2 # coding: utf-8
   3 # %%
   4 import itertools
   5 import collections
   6 import json
   7 import pandas as pd
   8 import numpy as np
   9 from scipy.stats import gmean
  10 import datetime
  11 import os
  12 import sqlalchemy
  13
  14 import matplotlib as mpl
  15 import matplotlib.pyplot as plt
  16 import matplotlib.animation as ani
  17 plt.ioff()
  18 # # %matplotlib inline
  19
  20
  21 # %%
  22 chart_start_date = '2020-09-15'
  23
  24
  25 # %%
  26 connection_string = 'postgresql://covid:3NbjJTkT63@localhost/covid'
  27
  28
  29 # %%
  30 engine = sqlalchemy.create_engine(connection_string)
  31
  32 # %%
  33 query_string = '''select uk_data.date,
  34     uk_data.new_cases, uk_data_7.new_cases as new_cases_7,
  35     uk_data.new_tests, uk_data_7.new_tests as new_tests_7,
  36     uk_data.new_pcr_tests, uk_data_7.new_pcr_tests as new_pcr_7,
  37     uk_data.new_pillar_1_2_tests as new_pillar_12, uk_data_7.new_pillar_1_2_tests as new_pillar_12_7,
  38      uk_data.new_cases::float / uk_data.new_tests as fraction_positive,
  39         uk_data_7.new_cases / uk_data_7.new_tests as fraction_positive_7
  40     from uk_data left outer join uk_data_7 using (date)
  41     order by uk_data.date'''
  42 tests_data = pd.read_sql_query(query_string, engine,
  43     index_col='date',
  44     parse_dates = ['date'])
  45
  46 # %%
  47 pri_y_max = int((tests_data.dropna().loc[chart_start_date: , 'new_tests_7'].max() * 1.1) / 100 ) * 100
  48 ax = tests_data.dropna().loc[chart_start_date: , 'new_tests_7'].plot(figsize=(10, 8),
  49                                                                style=['k-'],
  50                                                                legend=False,
  51                                                                ylim=(0, pri_y_max))
  52 ax.set_title('Tests done and new cases (7 day moving average)')
  53 ax.legend(['Tests, 7 day moving average'], loc='lower left')
  54 ax.set_ylabel('Tests')
  55 sec_y_max = int((tests_data.dropna().loc[chart_start_date:, 'new_cases_7'].max() * 1.1) / 100) * 100
  56 ax = tests_data.dropna().loc[chart_start_date:, 'new_cases_7'].plot(ax=ax, secondary_y=True, style='r--')
  57 ax.set_ylim((0, sec_y_max))
  58 ax.legend(['Cases (7 day moving average)'], loc='lower right')
  59 ax.set_ylabel('New cases')
  60 plt.savefig('tests_and_cases.png')
  61
  62
  63 # %%
  64 ax = (tests_data.loc[chart_start_date: , ['fraction_positive', 'fraction_positive_7']] * 100).plot(figsize=(10, 8),
  65                                                                                                   style=['b:', 'k-'], legend=False)
  66 ax.set_title('Fraction of tests with positive results')
  67 ax.legend(['Fraction positive (%)', 'Fraction positive (%), 7 day moving average'], loc='upper left')
  68 ax.set_ylabel('Fraction positive')
  69 cases_axis_max = (1.0 * tests_data.loc[chart_start_date:].new_cases_7.max()
  70                   * tests_data.loc[chart_start_date:].fraction_positive.max()
  71                   / tests_data.loc[chart_start_date:].fraction_positive_7.max()
  72                   )
  73
  74 ax2 = ax.twinx()
  75 ax2 = tests_data.loc[chart_start_date:, 'new_cases_7'].plot(ax=ax2, secondary_y=True, style='r--')
  76 ax2.set_ylim(-1000, cases_axis_max)
  77 ax2.legend(['Cases (7 day moving average)'], loc='center left')
  78 ax2.set_ylabel('New cases')
  79 plt.savefig('fraction_positive_tests.png')
  80
  81
  82 # %%
  83 ax = tests_data.dropna().loc[chart_start_date:].plot(x='fraction_positive_7', y='new_tests_7',
  84                                                   figsize=(8, 8),
  85                                                   legend=None)
  86 ax.set_xlabel("Fraction of tests that are positive")
  87 ax.set_ylabel("Number of tests")
  88 for d in tests_data.dropna().loc[chart_start_date::15].index:
  89     ax.plot(tests_data.loc[d, 'fraction_positive_7'], tests_data.loc[d, 'new_tests_7'], 'o',
  90                         markersize=8)#, markerfacecolor=marker_col, markeredgecolor=marker_col)
  91     ax.text(tests_data.loc[d, 'fraction_positive_7'] + 0.0002, tests_data.loc[d, 'new_tests_7'],
  92             s = d.strftime("%d %B %Y"))
  93 plt.savefig('fraction_positive_tests_vs_tests.png')
  94
  95
  96 # %%
  97 fig = plt.figure(figsize=(8, 8))
  98 plt.ylabel('Number of tests')
  99 plt.xlabel('Fraction of tests that are positive')
 100 all_data = tests_data.dropna().loc[chart_start_date:]
 101
 102
 103 minx = all_data.fraction_positive_7.min() * 0.9
 104 maxx = all_data.fraction_positive_7.max() * 1.1
 105 miny = all_data.new_tests_7.min() * 0.9
 106 maxy = all_data.new_tests_7.max() * 1.1
 107
 108 plt.xlim(minx, maxx)
 109 plt.ylim(miny, maxy)
 110 # plt.legend(None)
 111
 112 def build_state_frame(i):
 113     this_data = all_data[:i]
 114     plt.clf()
 115     plt.ylabel('Number of tests')
 116     plt.xlabel('Fraction of tests that are positive')
 117     plt.xlim(minx, maxx)
 118     plt.ylim(miny, maxy)
 119     p = plt.plot(this_data.fraction_positive_7, this_data.new_tests_7)
 120     p[0].set_color('r')
 121     for d in this_data[::15].index:
 122         plt.plot(this_data.loc[d, 'fraction_positive_7'],
 123                 this_data.loc[d, 'new_tests_7'], 'o',
 124                 markersize=8, markerfacecolor='r', markeredgecolor='r')
 125         plt.text(this_data.loc[d, 'fraction_positive_7'] + 0.0002,
 126                 this_data.loc[d, 'new_tests_7'],
 127             s = d.strftime("%d %B %Y"))
 128
 129 animator = ani.FuncAnimation(fig, build_state_frame,
 130                              frames=all_data.shape[0]+1,
 131                              interval=100,
 132                              repeat_delay=200,
 133                              repeat=True)
 134 animator.save('tests_vs_fraction_positive.mp4')
 135 # plt.show()
 136
 137
 138 # %%
 139 os.system('rm tests_vs_fraction_positive_animation.png')
 140 os.system('ffmpeg -i tests_vs_fraction_positive.mp4 -plays 0 -final_delay 1 -f apng tests_vs_fraction_positive_animation.png')
 141
 142
 143 # %%
 144
 145
 146
 147