def test_within_minus_one_part2(self):
large_data = self.data - 0.5
self.figs.append(
dcs.plot_correlation_matrix(large_data,
self.labels,
cmin=-1,
cmax=1))
def test_colormap(self):
self.figs.append(
dcs.plot_correlation_matrix(self.data, colormap='seismic_r'))
def test_within_minus_one(self):
large_data = self.data - 0.5
self.figs.append(dcs.plot_correlation_matrix(large_data, self.labels))
def test_above_one_part2(self):
large_data = self.data * 1000
self.figs.append(
dcs.plot_correlation_matrix(large_data, self.labels, cmax=2000))
def test_symmetric(self):
self.figs.append(dcs.plot_correlation_matrix(self.sym))
def test_type(self):
self.figs.append(
dcs.plot_correlation_matrix(self.data, type_=self.type_))
def test_nonsquare(self):
self.figs.append(dcs.plot_correlation_matrix(self.data[:5, :3], [self.labels[:3], \
self.labels[:5]]))
import numpy as np
import dstauffman as dcs
#%% Main function
if __name__ == '__main__':
#%% Create some fake data
# random data
data = np.random.rand(10, 10)
# normalize the random data
data = dcs.unit(data, axis=0)
# labels for the plot
labels = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j']
# make some symmetric data
sym = data.copy()
num = sym.shape[0]
for j in range(num):
for i in range(num):
if i == j:
sym[i, j] = 1
elif i > j:
sym[i, j] = data[j, i]
else:
pass
# create opts
opts = dcs.Opts()
#%% Create the plots
fig1 = dcs.plot_correlation_matrix(data, labels, opts=opts)
fig2 = dcs.plot_correlation_matrix(sym, labels, opts=opts)
def test_symmetric_all(self):
self.figs.append(dcs.plot_correlation_matrix(self.sym, plot_lower_only=False))
def test_symmetric(self):
self.figs.append(dcs.plot_correlation_matrix(self.sym))
def test_all_args(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, self.labels, self.type_, self.opts, \
matrix_name=self.matrix_name, cmin=0, cmax=1, colormap='viridis', xlabel='', ylabel='', \
plot_lower_only=False, label_values=True, x_lab_rot=180))
def test_type(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, type_=self.type_))
def test_default_labels(self):
self.figs.append(dcs.plot_correlation_matrix(self.data[:5, :3]))
def test_nonsquare(self):
self.figs.append(dcs.plot_correlation_matrix(self.data[:5, :3], [self.labels[:3], \
self.labels[:5]]))
def test_normal(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, self.labels))
def test_x_label_rotation(self):
self.figs.append(
dcs.plot_correlation_matrix(self.data, self.labels, x_lab_rot=0))
def test_bad_labels(self):
with self.assertRaises(ValueError):
self.figs.append(dcs.plot_correlation_matrix(self.data, ['a']))
def test_above_one_part2(self):
large_data = self.data * 1000
self.figs.append(dcs.plot_correlation_matrix(large_data, self.labels, cmax=2000))
def test_normal(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, self.labels))
def test_below_one_part2(self):
large_data = 1000*(self.data - 0.5)
self.figs.append(dcs.plot_correlation_matrix(large_data, self.labels, cmin=-2))
def test_default_labels(self):
self.figs.append(dcs.plot_correlation_matrix(self.data[:5, :3]))
def test_within_minus_one(self):
large_data = self.data - 0.5
self.figs.append(dcs.plot_correlation_matrix(large_data, self.labels))
def test_all_args(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, self.labels, self.type_, self.opts, \
matrix_name=self.matrix_name, cmin=0, cmax=1, colormap='viridis', xlabel='', ylabel='', \
plot_lower_only=False, label_values=True, x_lab_rot=180))
def test_within_minus_one_part2(self):
large_data = self.data - 0.5
self.figs.append(dcs.plot_correlation_matrix(large_data, self.labels, cmin=-1, cmax=1))
def test_symmetric_all(self):
self.figs.append(
dcs.plot_correlation_matrix(self.sym, plot_lower_only=False))
def test_colormap(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, colormap='seismic_r'))
def test_below_one_part2(self):
large_data = 1000 * (self.data - 0.5)
self.figs.append(
dcs.plot_correlation_matrix(large_data, self.labels, cmin=-2))
def test_ylabel(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, ylabel='Testing Label'))
def test_x_label_rotation(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, self.labels, x_lab_rot=0))
def test_nans(self):
self.data[0, 0] = np.nan
self.figs.append(dcs.plot_correlation_matrix(self.data, self.labels))
def test_ylabel(self):
self.figs.append(
dcs.plot_correlation_matrix(self.data, ylabel='Testing Label'))
def test_bad_labels(self):
with self.assertRaises(ValueError):
self.figs.append(dcs.plot_correlation_matrix(self.data, ['a']))
def test_nans(self):
self.data[0, 0] = np.nan
self.figs.append(dcs.plot_correlation_matrix(self.data, self.labels))
def test_label_values(self):
self.figs.append(dcs.plot_correlation_matrix(self.data, label_values=True))
def test_label_values(self):
self.figs.append(
dcs.plot_correlation_matrix(self.data, label_values=True))
import numpy as np
import dstauffman as dcs
#%% Main function
if __name__=='__main__':
#%% Create some fake data
# random data
data = np.random.rand(10, 10)
# normalize the random data
data = dcs.unit(data, axis=0)
# labels for the plot
labels = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j']
# make some symmetric data
sym = data.copy()
num = sym.shape[0]
for j in range(num):
for i in range(num):
if i == j:
sym[i, j] = 1
elif i > j:
sym[i, j] = data[j, i]
else:
pass
# create opts
opts = dcs.Opts()
#%% Create the plots
fig1 = dcs.plot_correlation_matrix(data, labels, opts=opts)
fig2 = dcs.plot_correlation_matrix(sym, labels, opts=opts)
