enable black for mobile-vision

Summary:
https://fb.workplace.com/groups/pythonfoundation/posts/2990917737888352

Remove `mobile-vision` from opt-out list; leaving `mobile-vision/SNPE` opted out because of 3rd-party code.

arc lint --take BLACK --apply-patches --paths-cmd 'hg files mobile-vision'

allow-large-files

Reviewed By: sstsai-adl

Differential Revision: D30721093

fbshipit-source-id: 9e5c16d988b315b93a28038443ecfb92efd18ef8

tests/misc/test_flop_count.py

+import os
 import tempfile
-from d2go.utils.testing.rcnn_helper import RCNNBaseTestCases
+
 from d2go.utils.flop_calculator import dump_flops_info
 from d2go.utils.testing.data_loader_helper import create_fake_detection_data_loader
-import os
+from d2go.utils.testing.rcnn_helper import RCNNBaseTestCases
 
 
 class TestFlopCount(RCNNBaseTestCases.TemplateTestCase):
@@ -19,6 +20,10 @@ class TestFlopCount(RCNNBaseTestCases.TemplateTestCase):
         with tempfile.TemporaryDirectory(prefix="d2go_test") as output_dir:
             dump_flops_info(self.test_model, inputs, output_dir)
 
-            for fname in ["flops_str_mobilecv", "flops_str_fvcore", "flops_table_fvcore"]:
+            for fname in [
+                "flops_str_mobilecv",
+                "flops_str_fvcore",
+                "flops_table_fvcore",
+            ]:
                 outf = os.path.join(output_dir, fname + ".txt")
                 self.assertTrue(os.path.isfile(outf))

tests/misc/test_utils_abnormal_checker.py

@@ -2,9 +2,9 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
 
 
+import glob
 import tempfile
 import unittest
-import glob
 
 import d2go.utils.abnormal_checker as ac
 import torch
@@ -35,9 +35,7 @@ class TestUtilsAbnormalChecker(unittest.TestCase):
         model = Model()
 
         with tempfile.TemporaryDirectory() as tmp_dir:
-            checker = ac.AbnormalLossChecker(
-                -1, writers=[ac.FileWriter(tmp_dir)]
-            )
+            checker = ac.AbnormalLossChecker(-1, writers=[ac.FileWriter(tmp_dir)])
             cmodel = ac.AbnormalLossCheckerWrapper(model, checker)
 
             losses = [5, 4, 3, 10, 9, 2, 5, 4]

tests/modeling/test_box2box_transform.py

@@ -2,16 +2,16 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
 
 
-import numpy as np
-import torch
 import unittest
 
+import numpy as np
+import torch
 from detectron2.modeling.box_regression import Box2BoxTransform
 
 
 class TestBox2BoxTransform(unittest.TestCase):
     def test_box2box_transform(self):
-        """ Match unit test UtilsBoxesTest.TestBboxTransformRandom in
+        """Match unit test UtilsBoxesTest.TestBboxTransformRandom in
         caffe2/operators/generate_proposals_op_util_boxes_test.cc
         """
         box2box_transform = Box2BoxTransform(weights=(1.0, 1.0, 1.0, 1.0))

tests/modeling/test_box_with_nms_limit.py

@@ -2,11 +2,11 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
 
 
-import torch
 import unittest
 
-from detectron2.modeling.roi_heads.fast_rcnn import fast_rcnn_inference
+import torch
 from detectron2.layers import cat
+from detectron2.modeling.roi_heads.fast_rcnn import fast_rcnn_inference
 from detectron2.structures import Boxes
 
 
@@ -37,10 +37,12 @@ class TestBoxWithNMSLimit(unittest.TestCase):
         scores = [torch.rand(n, num_class + 1, generator=rng) for n in batch_splits]
 
         ref_results, ref_kept_indices = fast_rcnn_inference(
-            boxes, scores, image_shapes,
+            boxes,
+            scores,
+            image_shapes,
             score_thresh=score_thresh,
             nms_thresh=nms_thresh,
-            topk_per_image=detections_per_im
+            topk_per_image=detections_per_im,
         )
         for result, kept_index, score in zip(ref_results, ref_kept_indices, scores):
             torch.testing.assert_allclose(
@@ -78,7 +80,14 @@ class TestBoxWithNMSLimit(unittest.TestCase):
             output_classes_include_bg_cls=False,
             legacy_plus_one=False,
         )
-        roi_score_nms, roi_bbox_nms, roi_class_nms, roi_batch_splits_nms, roi_keeps_nms, roi_keeps_size_nms = nms_outputs  # noqa
+        (
+            roi_score_nms,
+            roi_bbox_nms,
+            roi_class_nms,
+            roi_batch_splits_nms,
+            roi_keeps_nms,
+            roi_keeps_size_nms,
+        ) = nms_outputs  # noqa
 
         roi_score_nms = roi_score_nms.split(roi_batch_splits_nms.int().tolist())
         roi_bbox_nms = roi_bbox_nms.split(roi_batch_splits_nms.int().tolist())

tests/modeling/test_model_zoo.py

@@ -3,7 +3,8 @@ import unittest
 
 from d2go.model_zoo import model_zoo
 
-OSSRUN = os.getenv('OSSRUN') == '1'
+OSSRUN = os.getenv("OSSRUN") == "1"
+
 
 class TestD2GoModelZoo(unittest.TestCase):
     @unittest.skipIf(not OSSRUN, "OSS test only")
@@ -12,5 +13,6 @@ class TestD2GoModelZoo(unittest.TestCase):
         for cfgfile in configs:
             model = model_zoo.get(cfgfile, trained=True)
 
+
 if __name__ == "__main__":
     unittest.main()

tests/modeling/test_modeling_image_pooler.py

@@ -67,7 +67,7 @@ class TestModelingImagePooler(unittest.TestCase):
         self.assertArrayEqual(pooled_box, [[2, 3, 5, 7]])
 
     def test_image_pooler_scale_box_large_crop_only(self):
-        """ Crop bbox """
+        """Crop bbox"""
         H, W = 398, 224
         all_boxes = Boxes(torch.Tensor([[50, 40, 100, 80], [150, 60, 200, 120]]))
         image = rh.get_batched_inputs(1, (H, W), (H, W), all_boxes)[0]["image"]
@@ -86,7 +86,7 @@ class TestModelingImagePooler(unittest.TestCase):
         self.assertArrayEqual(sub_boxes.tensor, [[0, 0, 50, 40], [100, 20, 150, 80]])
 
     def test_image_pooler_scale_box_large_crop_and_scale(self):
-        """ Crop bbox that is scaled """
+        """Crop bbox that is scaled"""
         H, W = 398, 224
         all_boxes = Boxes(torch.Tensor([[50, 40, 100, 80], [150, 60, 200, 120]]))
         image = rh.get_batched_inputs(1, (H, W), (H, W), all_boxes)[0]["image"]
@@ -111,7 +111,7 @@ class TestModelingImagePooler(unittest.TestCase):
         self.assertArrayEqual(sub_boxes.tensor, [[15, 8, 65, 48], [115, 28, 165, 88]])
 
     def test_image_pooler_scale_box_large_crop_scale_and_resize(self):
-        """ Crop bbox that is scaled, resize the cropped box """
+        """Crop bbox that is scaled, resize the cropped box"""
         H, W = 398, 224
         all_boxes = Boxes(torch.Tensor([[50, 40, 100, 80], [150, 60, 200, 120]]))
         image = rh.get_batched_inputs(1, (H, W), (H, W), all_boxes)[0]["image"]

tests/modeling/test_nms.py

@@ -11,7 +11,7 @@ from detectron2.layers import nms as box_nms
 
 class TestNMS(unittest.TestCase):
     def test_nms_cpu(self):
-        """ Match unit test UtilsNMSTest.TestNMS in
+        """Match unit test UtilsNMSTest.TestNMS in
         caffe2/operators/generate_proposals_op_util_nms_test.cc
         """
 
@@ -60,7 +60,7 @@ class TestNMS(unittest.TestCase):
             np.testing.assert_array_equal(keep_indices, np.array(gt_index))
 
     def test_nms1_cpu(self):
-        """ Match unit test UtilsNMSTest.TestNMS1 in
+        """Match unit test UtilsNMSTest.TestNMS1 in
         caffe2/operators/generate_proposals_op_util_nms_test.cc
         """
 

tests/modeling/test_optimizer.py

@@ -2,10 +2,12 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
 
 
-import torch
 import unittest
-from d2go.optimizer import build_optimizer_mapper
+
 import d2go.runner.default_runner as default_runner
+import torch
+from d2go.optimizer import build_optimizer_mapper
+
 
 class TestArch(torch.nn.Module):
     def __init__(self):
@@ -22,6 +24,7 @@ class TestArch(torch.nn.Module):
         ret = self.avgpool(ret)
         return ret
 
+
 def _test_each_optimizer(cfg):
     model = TestArch()
     optimizer = build_optimizer_mapper(cfg, model)
@@ -33,12 +36,12 @@ def _test_each_optimizer(cfg):
         loss.backward()
         optimizer.step()
 
-class TestOptimizer(unittest.TestCase):
 
+class TestOptimizer(unittest.TestCase):
     def test_all_optimizers(self):
         runner = default_runner.Detectron2GoRunner()
         cfg = runner.get_default_cfg()
-        multipliers = [None, [{'conv': 0.1}]]
+        multipliers = [None, [{"conv": 0.1}]]
 
         for optimizer_name in ["SGD", "AdamW", "SGD_MT", "AdamW_MT"]:
             for mult in multipliers:

tests/modeling/test_rpn_heads.py

@@ -6,9 +6,9 @@ import logging
 import unittest
 
 import torch
+from d2go.runner import GeneralizedRCNNRunner
 from detectron2.modeling import build_anchor_generator, build_backbone
 from detectron2.modeling.proposal_generator import rpn
-from d2go.runner import GeneralizedRCNNRunner
 
 logger = logging.getLogger(__name__)
 
@@ -19,7 +19,7 @@ RPN_CFGS = {}
 
 class TestRPNHeads(unittest.TestCase):
     def test_build_rpn_heads(self):
-        """ Make sure rpn heads run """
+        """Make sure rpn heads run"""
 
         self.assertGreater(len(rpn.RPN_HEAD_REGISTRY._obj_map), 0)
 
@@ -65,7 +65,7 @@ class TestRPNHeads(unittest.TestCase):
                 )
 
     def test_build_rpn_heads_with_rotated_anchor_generator(self):
-        """ Make sure rpn heads work with rotated anchor generator"""
+        """Make sure rpn heads work with rotated anchor generator"""
 
         self.assertGreater(len(rpn.RPN_HEAD_REGISTRY._obj_map), 0)
 

tools/evaluator.py

@@ -6,9 +6,9 @@ Binary to evaluate predictor-based model (consist of models in deployable format
 torchscript, caffe2, etc.) using Detectron2Go system (dataloading, evaluation, etc).
 """
 
-import torch
 import logging
 
+import torch
 from d2go.distributed import launch
 from d2go.setup import (
     basic_argument_parser,

tools/lightning_train_net.py

@@ -41,7 +41,7 @@ class TrainOutput:
 
 
 def maybe_override_output_dir(cfg: CfgNode, output_dir: Optional[str]) -> None:
-    """Overrides the output directory if `output_dir` is not None. """
+    """Overrides the output directory if `output_dir` is not None."""
     if output_dir is not None and output_dir != cfg.OUTPUT_DIR:
         cfg.OUTPUT_DIR = output_dir
         logger.warning(
@@ -70,16 +70,21 @@ def _get_trainer_callbacks(cfg: CfgNode) -> List[Callback]:
         callbacks.append(QuantizationAwareTraining.from_config(cfg))
     return callbacks
 
+
 def _get_accelerator(use_cpu: bool) -> str:
     return "ddp_cpu" if use_cpu else "ddp"
 
 
-def get_trainer_params(cfg: CfgNode, num_machines: int, num_processes: int) -> Dict[str, Any]:
+def get_trainer_params(
+    cfg: CfgNode, num_machines: int, num_processes: int
+) -> Dict[str, Any]:
     use_cpu = cfg.MODEL.DEVICE.lower() == "cpu"
     accelerator = _get_accelerator(use_cpu)
     plugins = []
     if accelerator:
-        plugins.append(DDPPlugin(find_unused_parameters=cfg.MODEL.DDP_FIND_UNUSED_PARAMETERS))
+        plugins.append(
+            DDPPlugin(find_unused_parameters=cfg.MODEL.DDP_FIND_UNUSED_PARAMETERS)
+        )
 
     return {
         # training loop is bounded by max steps, use a large max_epochs to make
@@ -102,6 +107,7 @@ def get_trainer_params(cfg: CfgNode, num_machines: int, num_processes: int) -> D
         "plugins": plugins,
     }
 
+
 def do_train(
     cfg: CfgNode, trainer: pl.Trainer, task: GeneralizedRCNNTask
 ) -> Dict[str, str]: