def plot_gadf(ts, image_size, overlapping=False, scale='-1',
cmap='rainbow', output_file=None, interpolation=None):
"""Plot the image obtained after GADF transformation.
Parameters
----------
ts : np.array, shape = [n_features]
time series to plot
image_size : int (default = 32)
determines the shapes of the output image :
image_size x image_size
overlapping : bool (default = False)
if True, reducing the size of the time series with PAA is
done with possible overlapping windows.
scale : str (default = '-1')
the lower bound of the scaled time series. Possible values:
- '-1' : the time series are scaled in [-1,1]
- '0' : the time series are scaled in [0,1]
output_file : str or None (default = None)
if str, save the figure.
"""
# Check input data
if not (isinstance(ts, np.ndarray) and ts.ndim == 1):
raise ValueError("'ts' must be a 1-dimensional np.ndarray.")
# Size of ts
ts_size = ts.size
# Check parameters
if not isinstance(image_size, int):
raise TypeError("'image_size' must be an integer.")
if image_size < 2:
raise ValueError("'image_size' must be greater or equal than 2.")
if image_size > ts_size:
raise ValueError("'image_size' must be lower or equal than the size of 'ts'.")
if not isinstance(overlapping, (float, int)):
raise TypeError("'overlapping' must be a boolean.")
if scale not in ['0', '-1']:
raise ValueError("'scale' must be either '0' or '-1'.")
image_gadf = gaf(ts, ts_size, image_size, overlapping, 'd', scale)
plt.imshow(image_gadf, cmap=cmap, interpolation=interpolation)
plt.axis('off')
if output_file is not None:
plt.savefig(output_file)
# Show plot
plt.show()
def test_gaf():
"""Testing 'gaf'"""
# Parameter
size = 9
ts = np.linspace(-1, 1, size)
overlapping = False
ones = np.ones(size)
# Test 1
arr_actual = gaf(ts, size, size, overlapping, method='s', scale='-1')
arr_desired = np.outer(ts, ts) - np.outer(np.sqrt(ones - ts**2),
np.sqrt(ones - ts**2))
np.testing.assert_allclose(arr_actual, arr_desired, atol=1e-5, rtol=0.)
# Test 2
arr_actual = gaf(ts, size, size, overlapping, method='d', scale='-1')
arr_desired = np.outer(np.sqrt(ones - ts**2), ts) - np.outer(
ts, np.sqrt(ones - ts**2))
np.testing.assert_allclose(arr_actual, arr_desired, atol=1e-5, rtol=0.)