Reformat use yapf 0.27.0 & Add yapf style file (#672)

* Reformat use yapf 0.27.0

* Add yapf style file

mmdet/models/detectors/cascade_rcnn.py

@@ -186,8 +186,8 @@ class CascadeRCNN(BaseDetector, RPNTestMixin):
                                                 gt_labels, rcnn_train_cfg)
             loss_bbox = bbox_head.loss(cls_score, bbox_pred, *bbox_targets)
             for name, value in loss_bbox.items():
-                losses['s{}.{}'.format(
-                    i, name)] = (value * lw if 'loss' in name else value)
+                losses['s{}.{}'.format(i, name)] = (
+                    value * lw if 'loss' in name else value)
 
             # mask head forward and loss
             if self.with_mask:
@@ -224,8 +224,8 @@ class CascadeRCNN(BaseDetector, RPNTestMixin):
                     [res.pos_gt_labels for res in sampling_results])
                 loss_mask = mask_head.loss(mask_pred, mask_targets, pos_labels)
                 for name, value in loss_mask.items():
-                    losses['s{}.{}'.format(
-                        i, name)] = (value * lw if 'loss' in name else value)
+                    losses['s{}.{}'.format(i, name)] = (
+                        value * lw if 'loss' in name else value)
 
             # refine bboxes
             if i < self.num_stages - 1:
@@ -286,7 +286,8 @@ class CascadeRCNN(BaseDetector, RPNTestMixin):
                             [] for _ in range(mask_head.num_classes - 1)
                         ]
                     else:
-                        _bboxes = (det_bboxes[:, :4] * scale_factor
+                        _bboxes = (
+                            det_bboxes[:, :4] * scale_factor
                             if rescale else det_bboxes)
                         mask_rois = bbox2roi([_bboxes])
                         mask_feats = mask_roi_extractor(
@@ -324,7 +325,8 @@ class CascadeRCNN(BaseDetector, RPNTestMixin):
                     [] for _ in range(self.mask_head[-1].num_classes - 1)
                 ]
             else:
-                _bboxes = (det_bboxes[:, :4] * scale_factor
+                _bboxes = (
+                    det_bboxes[:, :4] * scale_factor
                     if rescale else det_bboxes)
                 mask_rois = bbox2roi([_bboxes])
                 aug_masks = []

mmdet/models/detectors/htc.py

@@ -224,8 +224,8 @@ class HybridTaskCascade(CascadeRCNN):
             roi_labels = bbox_targets[0]
 
             for name, value in loss_bbox.items():
-                losses['s{}.{}'.format(
-                    i, name)] = (value * lw if 'loss' in name else value)
+                losses['s{}.{}'.format(i, name)] = (
+                    value * lw if 'loss' in name else value)
 
             # mask head forward and loss
             if self.with_mask:
@@ -253,8 +253,8 @@ class HybridTaskCascade(CascadeRCNN):
                                                      gt_masks, rcnn_train_cfg,
                                                      semantic_feat)
                 for name, value in loss_mask.items():
-                    losses['s{}.{}'.format(
-                        i, name)] = (value * lw if 'loss' in name else value)
+                    losses['s{}.{}'.format(i, name)] = (
+                        value * lw if 'loss' in name else value)
 
             # refine bboxes (same as Cascade R-CNN)
             if i < self.num_stages - 1 and not self.interleaved:
@@ -312,7 +312,8 @@ class HybridTaskCascade(CascadeRCNN):
                             [] for _ in range(mask_head.num_classes - 1)
                         ]
                     else:
-                        _bboxes = (det_bboxes[:, :4] * scale_factor
+                        _bboxes = (
+                            det_bboxes[:, :4] * scale_factor
                             if rescale else det_bboxes)
                         mask_pred = self._mask_forward_test(
                             i, x, _bboxes, semantic_feat=semantic_feat)
@@ -345,7 +346,8 @@ class HybridTaskCascade(CascadeRCNN):
                     [] for _ in range(self.mask_head[-1].num_classes - 1)
                 ]
             else:
-                _bboxes = (det_bboxes[:, :4] * scale_factor
+                _bboxes = (
+                    det_bboxes[:, :4] * scale_factor
                     if rescale else det_bboxes)
 
                 mask_rois = bbox2roi([_bboxes])

mmdet/models/detectors/test_mixins.py

@@ -105,8 +105,8 @@ class MaskTestMixin(object):
         else:
             # if det_bboxes is rescaled to the original image size, we need to
             # rescale it back to the testing scale to obtain RoIs.
-            _bboxes = (det_bboxes[:, :4] * scale_factor
-                       if rescale else det_bboxes)
+            _bboxes = (
+                det_bboxes[:, :4] * scale_factor if rescale else det_bboxes)
             mask_rois = bbox2roi([_bboxes])
             mask_feats = self.mask_roi_extractor(
                 x[:len(self.mask_roi_extractor.featmap_strides)], mask_rois)

mmdet/models/mask_heads/fcn_mask_head.py

@@ -43,8 +43,8 @@ class FCNMaskHead(nn.Module):
 
         self.convs = nn.ModuleList()
         for i in range(self.num_convs):
-            in_channels = (self.in_channels
-                           if i == 0 else self.conv_out_channels)
+            in_channels = (
+                self.in_channels if i == 0 else self.conv_out_channels)
             padding = (self.conv_kernel_size - 1) // 2
             self.convs.append(
                 ConvModule(
@@ -54,8 +54,8 @@ class FCNMaskHead(nn.Module):
                     padding=padding,
                     conv_cfg=conv_cfg,
                     norm_cfg=norm_cfg))
-        upsample_in_channels = (self.conv_out_channels
-                                if self.num_convs > 0 else in_channels)
+        upsample_in_channels = (
+            self.conv_out_channels if self.num_convs > 0 else in_channels)
         if self.upsample_method is None:
             self.upsample = None
         elif self.upsample_method == 'deconv':
@@ -69,9 +69,9 @@ class FCNMaskHead(nn.Module):
                 scale_factor=self.upsample_ratio, mode=self.upsample_method)
 
         out_channels = 1 if self.class_agnostic else self.num_classes
-        logits_in_channel = (self.conv_out_channels
-                             if self.upsample_method == 'deconv' else
-                             upsample_in_channels)
+        logits_in_channel = (
+            self.conv_out_channels
+            if self.upsample_method == 'deconv' else upsample_in_channels)
         self.conv_logits = nn.Conv2d(logits_in_channel, out_channels, 1)
         self.relu = nn.ReLU(inplace=True)
         self.debug_imgs = None

mmdet/models/mask_heads/fused_semantic_head.py

@@ -85,10 +85,7 @@ class FusedSemanticHead(nn.Module):
         for i, feat in enumerate(feats):
             if i != self.fusion_level:
                 feat = F.interpolate(
-                    feat,
-                    size=fused_size,
-                    mode='bilinear',
-                    align_corners=True)
+                    feat, size=fused_size, mode='bilinear', align_corners=True)
                 x += self.lateral_convs[i](feat)
 
         for i in range(self.num_convs):

mmdet/models/registry.py

@@ -22,8 +22,9 @@ class Registry(object):
             module (:obj:`nn.Module`): Module to be registered.
         """
         if not issubclass(module_class, nn.Module):
-            raise TypeError('module must be a child of nn.Module, but got {}'.
-                            format(module_class))
+            raise TypeError(
+                'module must be a child of nn.Module, but got {}'.format(
+                    module_class))
         module_name = module_class.__name__
         if module_name in self._module_dict:
             raise KeyError('{} is already registered in {}'.format(

mmdet/models/utils/conv_module.py

@@ -98,7 +98,8 @@ class ConvModule(nn.Module):
             warnings.warn('ConvModule has norm and bias at the same time')
 
         # build convolution layer
-        self.conv = build_conv_layer(conv_cfg,
+        self.conv = build_conv_layer(
+            conv_cfg,
             in_channels,
             out_channels,
             kernel_size,

mmdet/models/utils/norm.py

 import torch.nn as nn
 
-
 norm_cfg = {
     # format: layer_type: (abbreviation, module)
     'BN': ('bn', nn.BatchNorm2d),

mmdet/ops/dcn/setup.py

@@ -8,8 +8,8 @@ setup(
             'src/deform_conv_cuda.cpp',
             'src/deform_conv_cuda_kernel.cu',
         ]),
-        CUDAExtension('deform_pool_cuda', [
-            'src/deform_pool_cuda.cpp', 'src/deform_pool_cuda_kernel.cu'
-        ]),
+        CUDAExtension(
+            'deform_pool_cuda',
+            ['src/deform_pool_cuda.cpp', 'src/deform_pool_cuda_kernel.cu']),
     ],
     cmdclass={'build_ext': BuildExtension})

tools/analyze_logs.py

@@ -51,8 +51,8 @@ def plot_curve(log_dicts, args):
         for j, metric in enumerate(metrics):
             print('plot curve of {}, metric is {}'.format(
                 args.json_logs[i], metric))
-            assert metric in log_dict[
-                epochs[0]], '{} does not contain metric {}'.format(
+            assert metric in log_dict[epochs[
+                0]], '{} does not contain metric {}'.format(
                     args.json_logs[i], metric)
 
             if 'mAP' in metric:

tools/test.py

@@ -63,15 +63,12 @@ def collect_results(result_part, size, tmpdir=None):
     if tmpdir is None:
         MAX_LEN = 512
         # 32 is whitespace
-        dir_tensor = torch.full((MAX_LEN, ),
-                                32,
-                                dtype=torch.uint8,
-                                device='cuda')
+        dir_tensor = torch.full(
+            (MAX_LEN, ), 32, dtype=torch.uint8, device='cuda')
         if rank == 0:
             tmpdir = tempfile.mkdtemp()
-            tmpdir = torch.tensor(bytearray(tmpdir.encode()),
-                                  dtype=torch.uint8,
-                                  device='cuda')
+            tmpdir = torch.tensor(
+                bytearray(tmpdir.encode()), dtype=torch.uint8, device='cuda')
             dir_tensor[:len(tmpdir)] = tmpdir
         dist.broadcast(dir_tensor, 0)
         tmpdir = dir_tensor.cpu().numpy().tobytes().decode().rstrip()
@@ -113,7 +110,8 @@ def parse_args():
         help='eval types')
     parser.add_argument('--show', action='store_true', help='show results')
     parser.add_argument('--tmpdir', help='tmp dir for writing some results')
-    parser.add_argument('--launcher',
+    parser.add_argument(
+        '--launcher',
         choices=['none', 'pytorch', 'slurm', 'mpi'],
         default='none',
         help='job launcher')
@@ -145,7 +143,8 @@ def main():
     # build the dataloader
     # TODO: support multiple images per gpu (only minor changes are needed)
     dataset = get_dataset(cfg.data.test)
-    data_loader = build_dataloader(dataset,
+    data_loader = build_dataloader(
+        dataset,
         imgs_per_gpu=1,
         workers_per_gpu=cfg.data.workers_per_gpu,
         dist=distributed,

tools/train.py

@@ -77,7 +77,8 @@ def main():
         # save mmdet version, config file content and class names in
         # checkpoints as meta data
         cfg.checkpoint_config.meta = dict(
-            mmdet_version=__version__, config=cfg.text,
+            mmdet_version=__version__,
+            config=cfg.text,
             classes=train_dataset.CLASSES)
     # add an attribute for visualization convenience
     model.CLASSES = train_dataset.CLASSES