def test_mean_diff_plot():
# Seed the random number generator.
# This ensures that the results below are reproducible.
np.random.seed(11111)
m1 = np.random.random(20)
m2 = np.random.random(20)
fig = plt.figure()
ax = fig.add_subplot(111)
# basic test.
fig = mean_diff_plot(m1, m2, ax=ax)
plt.close(fig)
# Test with pandas Series.
p1 = pd.Series(m1)
p2 = pd.Series(m2)
fig = mean_diff_plot(p1, p2)
plt.close(fig)
# Test plotting on assigned axis.
fig, ax = plt.subplots(2)
mean_diff_plot(m1, m2, ax=ax[0])
plt.close(fig)
# Test the setting of confidence intervals.
fig = mean_diff_plot(m1, m2, sd_limit=0)
plt.close(fig)
# Test asethetic controls.
fig = mean_diff_plot(m1, m2, scatter_kwds={'color': 'green', 's': 10})
plt.close(fig)
fig = mean_diff_plot(m1, m2, mean_line_kwds={'color': 'green', 'lw': 5})
plt.close(fig)
fig = mean_diff_plot(m1,
m2,
limit_lines_kwds={
'color': 'green',
'lw': 5,
'ls': 'dotted'
})
plt.close(fig)
def test_mean_diff_plot():
# Seed the random number generator.
# This ensures that the results below are reproducible.
np.random.seed(11111)
m1 = np.random.random(20)
m2 = np.random.random(20)
fig = plt.figure()
ax = fig.add_subplot(111)
# basic test.
fig = mean_diff_plot(m1, m2, ax=ax)
plt.close(fig)
# Test with pandas Series.
p1 = pd.Series(m1)
p2 = pd.Series(m2)
fig = mean_diff_plot(p1, p2)
plt.close(fig)
# Test plotting on assigned axis.
fig, ax = plt.subplots(2)
mean_diff_plot(m1, m2, ax = ax[0])
plt.close(fig)
# Test the setting of confidence intervals.
fig = mean_diff_plot(m1, m2, sd_limit = 0)
plt.close(fig)
# Test asethetic controls.
fig = mean_diff_plot(m1, m2, scatter_kwds={'color':'green', 's':10})
plt.close(fig)
fig = mean_diff_plot(m1, m2, mean_line_kwds={'color':'green', 'lw':5})
plt.close(fig)
fig = mean_diff_plot(m1, m2, limit_lines_kwds={'color':'green',
'lw':5,
'ls':'dotted'})
plt.close(fig)
if __name__ == '__main__':
data_dir = Path('../data/simulated') / 'actual_plots'
data = load_data(data_dir, 'aggregation_effects_S_*.csv', index_col=0)
data['level'] = data['level'].map(level_mapper)
data = data[data['order'].isin(
'blocks snake_20 snake_40 snake_60 snake_80 snake'.split())]
data = data[data['function'].isin('Martin_et_al_2000'.split())]
data = data[data['level'].isin([
v for k, v in level_mapper.items()
if k in 'squares squares-ghetto ghetto-squares ghetto'.split()
])]
data = data[data['bandwidth'] == 500]
data = data[data['cell'] == 25]
ols_model = smf.ols('S_by_plot ~ S_by_page', data=data).fit()
print(ols_model.summary())
# data.plot(kind='scatter', x='S_by_page', y='S_by_plot', c='level')
data.plot(
x='S_by_page',
y='S_by_plot',
c='level',
kind='scatter',
cmap=cm.get_cmap('viridis', len(level_mapper)),
)
mean_diff_plot(data['S_by_plot'], data['S_by_page'], sd_limit=3)
plt.show()
ai_t1 = np.array(ai_t1)
jh_t1 = np.array(jh_t1)
hx_t1 = np.array(hx_t1)
fig, axes = plt.subplots(1, 3, sharey='row', figsize=(15, 6))
for i, (ax, title) in enumerate(
zip(axes, ('\nAI versus E1', '\nAI versus E2', '\nE1 versus E2'))):
ax.set_title(title, fontsize=16)
if i > 0:
ax.axes.get_yaxis().set_visible(False)
ax.xaxis.set_ticks(np.arange(0, 2000, 400))
ax.yaxis.set_ticks(np.arange(-800, 1800, 400))
mean_diff_plot(ai_t1, jh_t1, ax=axes[0], scatter_kwds={'alpha': 0.25})
mean_diff_plot(ai_t1, hx_t1, ax=axes[1], scatter_kwds={'alpha': 0.25})
mean_diff_plot(jh_t1, hx_t1, ax=axes[2], scatter_kwds={'alpha': 0.25})
for ax in axes:
ax.axvline(x=850, linestyle='--')
plt.setp(axes, xlim=(0, 1800))
plt.setp(axes, ylim=(-900, 900))
plt.subplots_adjust(wspace=0.1, hspace=0, left=0.075)
fig.suptitle(
'T1 (ms) - Bland-Altman plot comparing AI & expert measurements\n',
fontsize=20)
fig.show()
######