8185 test refactor 2

[garciadias]

Fixes #8185

Description

This PR solves items 2 and 3 on #8185 for a few test folders.
I would merge these and proceed with the same type of change in other files if @ericspod approves.

I would like to keep these PRs small, so even if they have the same pattern of changes, merging them bit by bit would make them more manageable.

A few sentences describing the changes proposed in this pull request.

Types of changes

• Non-breaking change (fix or new feature that would not break existing functionality).
• Breaking change (fix or new feature that would cause existing functionality to change).
• New tests added to cover the changes.
• Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
• Quick tests passed locally by running ./runtests.sh --quick --unittests --disttests.
• In-line docstrings updated.
• Documentation updated, tested make html command in the docs/ folder.

[garciadias]

Hi @ericspod,

Could you please review this PR?
I think it is all good to follow your specification on the #8185 issue.
I am sending the modification to these few tests so we can limit the size of the PR.
If you approve, I will proceed with more of these simplifications.

[ericspod]

If there's only one key passed to dict_product, this isn't saving any space or complexity so I wouldn't use this here.

[ericspod]

Suggested change

	for params in dict_product(device=TEST_DEVICES, dtype=DTYPES):
	TESTS.append((*params["device"], *params["dtype"])) # type: ignore
	TESTS = [(*p["device"], *p["dtype"]) for p in dict_product(device=TEST_DEVICES, dtype=DTYPES)] # type: ignore

In some places a list comprehension may be fine to use, and not have to have TESTS = [] before.

[ericspod]

Suggested change

	for params in dict_product(
	dropout_rate=[0.5],
	in_channels=[1, 4],
	hidden_size=[96, 288],
	img_size=[32, 64],
	patch_size=[8, 16],
	num_heads=[8, 12],
	proj_type=["conv", "perceptron"],
	pos_embed_type=["none", "learnable", "sincos"],
	nd=[2, 3],
	):
	test_case = [
	{
	"in_channels": params["in_channels"],
	"img_size": (params["img_size"],) * params["nd"],
	"patch_size": (params["patch_size"],) * params["nd"],
	"hidden_size": params["hidden_size"],
	"num_heads": params["num_heads"],
	"proj_type": params["proj_type"],
	"pos_embed_type": params["pos_embed_type"],
	"dropout_rate": params["dropout_rate"],
	"spatial_dims": params["nd"],
	},
	(2, params["in_channels"], *[params["img_size"]] * params["nd"]),
	(2, (params["img_size"] // params["patch_size"]) ** params["nd"], params["hidden_size"]),
	]
	TEST_CASE_PATCHEMBEDDINGBLOCK.append(test_case)
	for params in dict_product(
	dropout_rate=[0.5],
	in_channels=[1, 4],
	hidden_size=[96, 288],
	img_size=[32, 64],
	patch_size=[8, 16],
	num_heads=[8, 12],
	proj_type=["conv", "perceptron"],
	pos_embed_type=["none", "learnable", "sincos"],
	spatial_dims=[2, 3],
	):
	nd = params["spatial_dims"]
	args = {**params, "img_size": (params["img_size"],) * nd, "patch_size": (params["patch_size"],) * nd}
	input_shape = (2, params["in_channels"], *[params["img_size"]] * nd),
	expected_shape = (2, (params["img_size"] // params["patch_size"]) ** nd, params["hidden_size"]),
	TEST_CASE_PATCHEMBEDDINGBLOCK.append([args, input_shape, expected_shape])

With some changes to argument choices you can eliminate needing to make an argument dictionary by using the initial dictionary from dict_product. Elsewhere you can do the same buy may need to remove keys from args where they aren't desired. Same idea for the other changes in this file.