Solves path problem in test_bundle_trt_export.py

README.md

@@ -18,12 +18,13 @@
 
 MONAI is a [PyTorch](https://pytorch.org/)-based, [open-source](https://github.com/Project-MONAI/MONAI/blob/dev/LICENSE) framework for deep learning in healthcare imaging, part of the [PyTorch Ecosystem](https://pytorch.org/ecosystem/).
 Its ambitions are as follows:
+
 - Developing a community of academic, industrial and clinical researchers collaborating on a common foundation;
 - Creating state-of-the-art, end-to-end training workflows for healthcare imaging;
 - Providing researchers with the optimized and standardized way to create and evaluate deep learning models.
 
-
 ## Features
+
 > _Please see [the technical highlights](https://docs.monai.io/en/latest/highlights.html) and [What's New](https://docs.monai.io/en/latest/whatsnew.html) of the milestone releases._
 
 - flexible pre-processing for multi-dimensional medical imaging data;

tests/bundle/test_bundle_trt_export.py

@@ -70,7 +70,7 @@ def tearDown(self):
     @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_4])
     @unittest.skipUnless(has_torchtrt and has_tensorrt, "Torch-TensorRT is required for conversion!")
     def test_trt_export(self, convert_precision, input_shape, dynamic_batch):
-        tests_dir = Path(__file__).resolve().parent
+        tests_dir = Path(__file__).resolve().parents[1]
         meta_file = os.path.join(tests_dir, "testing_data", "metadata.json")
         config_file = os.path.join(tests_dir, "testing_data", "inference.json")
         with tempfile.TemporaryDirectory() as tempdir:

tests/networks/test_convert_to_onnx.py

@@ -64,7 +64,7 @@ def test_unet(self, device, use_trace, use_ort):
                 rtol=rtol,
                 atol=atol,
             )
-        self.assertTrue(isinstance(onnx_model, onnx.ModelProto))
+            self.assertTrue(isinstance(onnx_model, onnx.ModelProto))
 
     @parameterized.expand(TESTS_ORT)
     @SkipIfBeforePyTorchVersion((1, 12))

tests/test_utils.py

@@ -30,9 +30,10 @@
 import warnings
 from contextlib import contextmanager
 from functools import partial, reduce
+from itertools import product
 from pathlib import Path
 from subprocess import PIPE, Popen
-from typing import Callable
+from typing import Callable, Literal
 from urllib.error import ContentTooShortError, HTTPError
 
 import numpy as np
@@ -862,6 +863,21 @@ def equal_state_dict(st_1, st_2):
 if torch.cuda.is_available():
     TEST_DEVICES.append([torch.device("cuda")])
 
+
+def dict_product(trailing=False, format: Literal["list", "dict"] = "dict", **items):
+    keys = items.keys()
+    values = items.values()
+    for pvalues in product(*values):
+        dict_comb = dict(zip(keys, pvalues))
+        if format == "dict":
+            if trailing:
+                yield [dict_comb] + list(pvalues)
+            else:
+                yield dict_comb
+        else:
+            yield pvalues
+
+
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(prog="util")
     parser.add_argument("-c", "--count", default=2, help="max number of gpus")

tests/transforms/test_gibbs_noise.py

@@ -21,14 +21,12 @@
 from monai.transforms import GibbsNoise
 from monai.utils.misc import set_determinism
 from monai.utils.module import optional_import
-from tests.test_utils import TEST_NDARRAYS, assert_allclose
+from tests.test_utils import TEST_NDARRAYS, assert_allclose, dict_product
 
 _, has_torch_fft = optional_import("torch.fft", name="fftshift")
 
-TEST_CASES = []
-for shape in ((128, 64), (64, 48, 80)):
-    for input_type in TEST_NDARRAYS if has_torch_fft else [np.array]:
-        TEST_CASES.append((shape, input_type))
+params = {"shape": ((128, 64), (64, 48, 80)), "input_type": TEST_NDARRAYS if has_torch_fft else [np.array]}
+TEST_CASES = list(dict_product(format="list", **params))
 
 
 class TestGibbsNoise(unittest.TestCase):