Merge branch 'dev' into add-trt-support-for-bundles

.pre-commit-config.yaml

@@ -7,6 +7,8 @@ ci:
   autoupdate_schedule: quarterly
   # submodules: true
 
+exclude: '.*\.ipynb$'
+
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v4.4.0

ci/run_premerge_cpu.sh

@@ -39,6 +39,8 @@ remove_pipenv() {
     echo "removing pip environment"
     pipenv --rm
     rm Pipfile Pipfile.lock
+    pipenv --clear
+    df -h
 }
 
 verify_bundle() {

ci/unit_tests/test_endoscopic_tool_segmentation.py

@@ -0,0 +1,108 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import shutil
+import tempfile
+import unittest
+
+import numpy as np
+from monai.bundle import ConfigWorkflow
+from monai.data import PILWriter
+from parameterized import parameterized
+from utils import check_workflow
+
+TEST_CASE_1 = [  # train, evaluate
+    {
+        "bundle_root": "models/endoscopic_tool_segmentation",
+        "train#trainer#max_epochs": 2,
+        "train#dataloader#num_workers": 1,
+        "validate#dataloader#num_workers": 1,
+        "train#deterministic_transforms#3#spatial_size": [32, 32],
+    }
+]
+
+TEST_CASE_2 = [  # inference
+    {
+        "bundle_root": "models/endoscopic_tool_segmentation",
+        "handlers#0#_disabled_": True,
+        "preprocessing#transforms#2#spatial_size": [32, 32],
+    }
+]
+
+
+class TestEndoscopicSeg(unittest.TestCase):
+    def setUp(self):
+        self.dataset_dir = tempfile.mkdtemp()
+        dataset_size = 10
+        writer = PILWriter(np.uint8)
+        shape = (736, 480)
+        for mode in ["train", "val", "test"]:
+            for sub_folder in ["inbody", "outbody"]:
+                sample_dir = os.path.join(self.dataset_dir, f"{mode}/{sub_folder}")
+                os.makedirs(sample_dir)
+                for s in range(dataset_size):
+                    image = np.random.randint(low=0, high=5, size=(3, *shape)).astype(np.int8)
+                    image_filename = os.path.join(sample_dir, f"{sub_folder}_{s}.jpg")
+                    writer.set_data_array(image, channel_dim=0)
+                    writer.write(image_filename, verbose=True)
+                    if mode != "test":
+                        label = np.random.randint(low=0, high=5, size=shape).astype(np.int8)
+                        label_filename = os.path.join(sample_dir, f"{sub_folder}_{s}_seg.jpg")
+                        writer.set_data_array(label, channel_dim=None)
+                        writer.write(label_filename, verbose=True)
+
+    def tearDown(self):
+        shutil.rmtree(self.dataset_dir)
+
+    @parameterized.expand([TEST_CASE_1])
+    def test_train_eval_config(self, override):
+        override["dataset_dir"] = self.dataset_dir
+        bundle_root = override["bundle_root"]
+        train_file = os.path.join(bundle_root, "configs/train.json")
+        eval_file = os.path.join(bundle_root, "configs/evaluate.json")
+
+        trainer = ConfigWorkflow(
+            workflow="train",
+            config_file=train_file,
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            **override,
+        )
+        check_workflow(trainer, check_properties=True)
+
+        validator = ConfigWorkflow(
+            # override train.json, thus set the workflow to "train" rather than "eval"
+            workflow="train",
+            config_file=[train_file, eval_file],
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            **override,
+        )
+        check_workflow(validator, check_properties=True)
+
+    @parameterized.expand([TEST_CASE_2])
+    def test_infer_config(self, override):
+        override["dataset_dir"] = self.dataset_dir
+        bundle_root = override["bundle_root"]
+
+        inferrer = ConfigWorkflow(
+            workflow="infer",
+            config_file=os.path.join(bundle_root, "configs/inference.json"),
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            **override,
+        )
+        check_workflow(inferrer, check_properties=True)
+
+
+if __name__ == "__main__":
+    unittest.main()

ci/unit_tests/test_endoscopic_tool_segmentation_dist.py

@@ -0,0 +1,119 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import shutil
+import tempfile
+import unittest
+
+import numpy as np
+import torch
+from monai.data import PILWriter
+from parameterized import parameterized
+from utils import export_config_and_run_mgpu_cmd
+
+TEST_CASE_1 = [
+    {
+        "bundle_root": "models/endoscopic_tool_segmentation",
+        "train#trainer#max_epochs": 1,
+        "train#dataloader#num_workers": 1,
+        "validate#dataloader#num_workers": 1,
+        "train#deterministic_transforms#3#spatial_size": [32, 32],
+    }
+]
+
+TEST_CASE_2 = [
+    {
+        "bundle_root": "models/endoscopic_tool_segmentation",
+        "validate#dataloader#num_workers": 4,
+        "train#deterministic_transforms#3#spatial_size": [32, 32],
+    }
+]
+
+
+def test_order(test_name1, test_name2):
+    def get_order(name):
+        if "train" in name:
+            return 1
+        if "eval" in name:
+            return 2
+        if "infer" in name:
+            return 3
+        return 4
+
+    return get_order(test_name1) - get_order(test_name2)
+
+
+class TestEndoscopicSegMGPU(unittest.TestCase):
+    def setUp(self):
+        self.dataset_dir = tempfile.mkdtemp()
+        dataset_size = 10
+        writer = PILWriter(np.uint8)
+        shape = (3, 256, 256)
+        for mode in ["train", "val"]:
+            for sub_folder in ["inbody", "outbody"]:
+                sample_dir = os.path.join(self.dataset_dir, f"{mode}/{sub_folder}")
+                os.makedirs(sample_dir)
+                for s in range(dataset_size):
+                    image = np.random.randint(low=0, high=5, size=(3, *shape)).astype(np.int8)
+                    image_filename = os.path.join(sample_dir, f"{sub_folder}_{s}.jpg")
+                    writer.set_data_array(image, channel_dim=0)
+                    writer.write(image_filename, verbose=True)
+                    label = np.random.randint(low=0, high=5, size=shape).astype(np.int8)
+                    label_filename = os.path.join(sample_dir, f"{sub_folder}_{s}_seg.jpg")
+                    writer.set_data_array(label, channel_dim=None)
+                    writer.write(label_filename, verbose=True)
+
+    def tearDown(self):
+        shutil.rmtree(self.dataset_dir)
+
+    @parameterized.expand([TEST_CASE_1])
+    def test_train_mgpu_config(self, override):
+        override["dataset_dir"] = self.dataset_dir
+        bundle_root = override["bundle_root"]
+        train_file = os.path.join(bundle_root, "configs/train.json")
+        mgpu_train_file = os.path.join(bundle_root, "configs/multi_gpu_train.json")
+        output_path = os.path.join(bundle_root, "configs/train_override.json")
+        n_gpu = torch.cuda.device_count()
+        export_config_and_run_mgpu_cmd(
+            config_file=[train_file, mgpu_train_file],
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            override_dict=override,
+            output_path=output_path,
+            ngpu=n_gpu,
+            check_config=True,
+        )
+
+    @parameterized.expand([TEST_CASE_2])
+    def test_evaluate_mgpu_config(self, override):
+        override["dataset_dir"] = self.dataset_dir
+        bundle_root = override["bundle_root"]
+        train_file = os.path.join(bundle_root, "configs/train.json")
+        evaluate_file = os.path.join(bundle_root, "configs/evaluate.json")
+        mgpu_evaluate_file = os.path.join(bundle_root, "configs/multi_gpu_evaluate.json")
+        output_path = os.path.join(bundle_root, "configs/evaluate_override.json")
+        n_gpu = torch.cuda.device_count()
+        export_config_and_run_mgpu_cmd(
+            config_file=[train_file, evaluate_file, mgpu_evaluate_file],
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            override_dict=override,
+            output_path=output_path,
+            ngpu=n_gpu,
+            check_config=True,
+        )
+
+
+if __name__ == "__main__":
+    loader = unittest.TestLoader()
+    loader.sortTestMethodsUsing = test_order
+    unittest.main(testLoader=loader)

ci/unit_tests/test_lung_nodule_ct_detection.py

@@ -0,0 +1,126 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import json
+import os
+import shutil
+import sys
+import tempfile
+import unittest
+
+import nibabel as nib
+import numpy as np
+from monai.bundle import ConfigWorkflow
+from monai.data import create_test_image_3d
+from monai.utils import set_determinism
+from parameterized import parameterized
+from utils import check_workflow
+
+set_determinism(123)
+
+TEST_CASE_1 = [  # train, evaluate
+    {
+        "bundle_root": "models/lung_nodule_ct_detection",
+        "epochs": 3,
+        "batch_size": 1,
+        "val_interval": 2,
+        "train#dataloader#num_workers": 1,
+        "validate#dataloader#num_workers": 1,
+    }
+]
+
+TEST_CASE_2 = [{"bundle_root": "models/lung_nodule_ct_detection"}]  # inference
+
+
+def test_order(test_name1, test_name2):
+    def get_order(name):
+        if "train" in name:
+            return 1
+        if "eval" in name:
+            return 2
+        if "infer" in name:
+            return 3
+        return 4
+
+    return get_order(test_name1) - get_order(test_name2)
+
+
+class TestLungNoduleDetection(unittest.TestCase):
+    def setUp(self):
+        self.dataset_dir = tempfile.mkdtemp()
+        dataset_size = 3
+        train_patch_size = (192, 192, 80)
+        dataset_json = {}
+
+        img, _ = create_test_image_3d(train_patch_size[0], train_patch_size[1], train_patch_size[2], 2)
+        image_filename = os.path.join(self.dataset_dir, "image.nii.gz")
+        nib.save(nib.Nifti1Image(img, np.eye(4)), image_filename)
+        label = [0, 0]
+        box = [[108, 119, 131, 142, 26, 37], [132, 147, 149, 164, 25, 40]]
+        data = {"box": box, "image": image_filename, "label": label}
+        dataset_json["training"] = [data for _ in range(dataset_size)]
+        dataset_json["validation"] = [data for _ in range(dataset_size)]
+
+        self.ds_file = os.path.join(self.dataset_dir, "dataset.json")
+        with open(self.ds_file, "w") as fp:
+            json.dump(dataset_json, fp, indent=2)
+
+    def tearDown(self):
+        shutil.rmtree(self.dataset_dir)
+
+    @parameterized.expand([TEST_CASE_1])
+    def test_train_eval_config(self, override):
+        override["dataset_dir"] = self.dataset_dir
+        override["data_list_file_path"] = self.ds_file
+        bundle_root = override["bundle_root"]
+        train_file = os.path.join(bundle_root, "configs/train.json")
+        eval_file = os.path.join(bundle_root, "configs/evaluate.json")
+
+        sys.path.append(bundle_root)
+        trainer = ConfigWorkflow(
+            workflow="train",
+            config_file=train_file,
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            **override,
+        )
+        check_workflow(trainer, check_properties=False)
+
+        validator = ConfigWorkflow(
+            # override train.json, thus set the workflow to "train" rather than "eval"
+            workflow="train",
+            config_file=[train_file, eval_file],
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            **override,
+        )
+        check_workflow(validator, check_properties=False)
+
+    @parameterized.expand([TEST_CASE_2])
+    def test_infer_config(self, override):
+        override["dataset_dir"] = self.dataset_dir
+        override["data_list_file_path"] = self.ds_file
+        bundle_root = override["bundle_root"]
+
+        inferrer = ConfigWorkflow(
+            workflow="infer",
+            config_file=os.path.join(bundle_root, "configs/inference.json"),
+            logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+            meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+            **override,
+        )
+        check_workflow(inferrer, check_properties=True)
+
+
+if __name__ == "__main__":
+    loader = unittest.TestLoader()
+    loader.sortTestMethodsUsing = test_order
+    unittest.main(testLoader=loader)