[MNT] Test ElasticEnsemble with all distances

[MatthewMiddlehurst]

Describe the issue

The EE classifier does not test with all distances according to the coverage stats.

Suggest a potential alternative/fix

Write a test in test_elastic_ensemble.py which ensures this works correctly. Other parameters can be tweaked to speed up the test if necessary and it does not compromise the test output.

Additional context

No response

[CodeLionX]

Small addition: I guess the all-case should probably use aeon.distances.get_distance_function_names() instead of a hard-coded list of distances:

aeon/aeon/classification/distance_based/_elastic_ensemble.py

Lines 140 to 152 in 2dfca9c

	if self.distance_measures == "all":
	self._distance_measures = [
	"dtw",
	"ddtw",
	"wdtw",
	"wddtw",
	"lcss",
	"erp",
	"msm",
	"euclidean",
	"twe",
	]
	else:

The only reason not to do this is incompatible distances, but I don't know if this is the case.

[itsdivya1309]

Small addition: I guess the all-case should probably use aeon.distances.get_distance_function_names() instead of a hard-coded list of distances:

aeon/aeon/classification/distance_based/_elastic_ensemble.py

Lines 140 to 152 in 2dfca9c

	if self.distance_measures == "all":
	self._distance_measures = [
	"dtw",
	"ddtw",
	"wdtw",
	"wddtw",
	"lcss",
	"erp",
	"msm",
	"euclidean",
	"twe",
	]
	else:

The only reason not to do this is incompatible distances, but I don't know if this is the case.

Hi!
I think the reason for not using get_distance_function_names() is that the list will be appended whenever a new distance function is implemented. Then, EE would need to be updated every time, as not all distance measures are used in EE.

[CodeLionX]

Ok, then ignore my comment 😉