Reformatting of code

classifier/tools/train.py

@@ -63,9 +63,7 @@ def train_model_type(model, cfg, csv_path: str, json_path: str, box_dir: str, up
     lr_scheduler = optim.lr_scheduler.ReduceLROnPlateau(
         optimizer, mode='max', factor=0.1, patience=5, min_lr=1e-07, eps=1e-07, verbose=True)
 
-    dataloaders = {}
-    dataloaders['train'] = train_dataloader
-    dataloaders['val'] = val_dataloader
+    dataloaders = {'train': train_dataloader, 'val': val_dataloader}
 
     save_path = osp.join(cfg['save_path'], cfg['date'], cfg['type'])
     os.makedirs(save_path, exist_ok=True)

classifier/tools/visualize_prediction.py

@@ -72,7 +72,7 @@ def main_veh():
     print(f'Visualize predictions to {save_dir}')
 
     for key, val in cfg_veh["class_map"].items():
-        if (type(val) == int):
+        if type(val) == int:
             continue
         os.makedirs(osp.join(save_dir, val), exist_ok=True)
 

classifier/train_jeepsuv.py

@@ -57,9 +57,7 @@ def train_model_type(model, cfg, csv_path: str, json_path: str, box_dir: str):
     lr_scheduler = optim.lr_scheduler.ReduceLROnPlateau(
         optimizer, mode='min', factor=0.1, patience=10, min_lr=1e-07, eps=1e-07, verbose=True)
 
-    dataloaders = {}
-    dataloaders['train'] = train_dataloader
-    dataloaders['val'] = val_dataloader
+    dataloaders = {'train': train_dataloader, 'val': val_dataloader}
 
     save_path = osp.join(cfg['save_path'], cfg['date'], cfg['type'])
     os.makedirs(save_path, exist_ok=True)

detector/stop_detector.py

@@ -30,7 +30,7 @@ def process(self, track_data: dict):
         print(f'mean distance: {mean_distance}')
         print(distances)
         for i in range(N-self.skip_frame):
-            if (distances[i] < mean_distance*self.alpha):
+            if distances[i] < mean_distance*self.alpha:
                 return True
 
         return False

object_tracking/deepsort.py

@@ -6,7 +6,7 @@
 import numpy as np
 
 
-class deepsort_rbc():
+class deepsort_rbc:
 	def __init__(self):
 		self.metric = nn_matching.NearestNeighborDistanceMetric("cosine",.5 , 70)
 		self.tracker= Tracker(self.metric)
@@ -17,7 +17,7 @@ def reset_tracker(self):
 		self.tracker= Tracker(self.metric, max_age=12, n_init=1)			
 
 	def run_deep_sort(self, out_scores, out_boxes, features):
-		if out_boxes==[]:			
+		if not out_boxes:
 			self.tracker.predict()
 			print('No detections')
 			trackers = self.tracker.tracks

object_tracking/deepsort_feat.py

@@ -7,7 +7,7 @@
 import numpy as np
 
 
-class deepsort_rbc():
+class deepsort_rbc:
 	def __init__(self):
 		self.metric = nn_matching.NearestNeighborDistanceMetric("cosine",.4 , 70)
 		self.tracker= Tracker(self.metric)
@@ -18,7 +18,7 @@ def reset_tracker(self):
 		self.tracker= Tracker(self.metric, max_age=12, n_init=1)			
 
 	def run_deep_sort(self, out_scores, out_boxes, features):
-		if out_boxes==[]:			
+		if not out_boxes:
 			self.tracker.predict()
 			print('No detections')
 			trackers = self.tracker.tracks

object_tracking/reid/tools/parse_test_res.py

@@ -67,11 +67,11 @@ def parse_file(filepath, regex_mAP, regex_r1, regex_r5, regex_r10, regex_r20):
 
 
 def main(args):
-    regex_mAP = re.compile(r'mAP: ([\.\deE+-]+)%')
-    regex_r1 = re.compile(r'Rank-1  : ([\.\deE+-]+)%')
-    regex_r5 = re.compile(r'Rank-5  : ([\.\deE+-]+)%')
-    regex_r10 = re.compile(r'Rank-10 : ([\.\deE+-]+)%')
-    regex_r20 = re.compile(r'Rank-20 : ([\.\deE+-]+)%')
+    regex_mAP = re.compile(r'mAP: ([.\deE+-]+)%')
+    regex_r1 = re.compile(r'Rank-1 {2}: ([.\deE+-]+)%')
+    regex_r5 = re.compile(r'Rank-5 {2}: ([.\deE+-]+)%')
+    regex_r10 = re.compile(r'Rank-10 : ([.\deE+-]+)%')
+    regex_r20 = re.compile(r'Rank-20 : ([.\deE+-]+)%')
 
     final_res = defaultdict(list)
 

object_tracking/reid/torchreid/data/transforms.py

@@ -71,8 +71,10 @@ def __init__(
         sl=0.02,
         sh=0.4,
         r1=0.3,
-        mean=[0.4914, 0.4822, 0.4465]
+            mean=None
     ):
+        if mean is None:
+            mean = [0.4914, 0.4822, 0.4465]
         self.probability = probability
         self.mean = mean
         self.sl = sl
@@ -234,8 +236,8 @@ def build_transforms(
     height,
     width,
     transforms='random_flip',
-    norm_mean=[0.485, 0.456, 0.406],
-    norm_std=[0.229, 0.224, 0.225],
+        norm_mean=None,
+        norm_std=None,
     **kwargs
 ):
     """Builds train and test transform functions.
@@ -249,6 +251,10 @@ def build_transforms(
         norm_std (list or None, optional): normalization standard deviation values. Default is
             ImageNet standard deviation values.
     """
+    if norm_mean is None:
+        norm_mean = [0.485, 0.456, 0.406]
+    if norm_std is None:
+        norm_std = [0.229, 0.224, 0.225]
     if transforms is None:
         transforms = []
 

object_tracking/reid/torchreid/engine/engine.py

@@ -126,7 +126,7 @@ def run(
         visrank=False,
         visrank_topk=10,
         use_metric_cuhk03=False,
-        ranks=[1, 5, 10, 20],
+            ranks=None,
         rerank=False
     ):
         r"""A unified pipeline for training and evaluating a model.
@@ -160,6 +160,8 @@ def run(
                 Default is False. This is only enabled when test_only=True.
         """
 
+        if ranks is None:
+            ranks = [1, 5, 10, 20]
         if visrank and not test_only:
             raise ValueError(
                 'visrank can be set to True only if test_only=True'
@@ -297,7 +299,7 @@ def test(
         visrank_topk=10,
         save_dir='',
         use_metric_cuhk03=False,
-        ranks=[1, 5, 10, 20],
+            ranks=None,
         rerank=False
     ):
         r"""Tests model on target datasets.
@@ -313,6 +315,8 @@ def test(
             ``extract_features()`` and ``parse_data_for_eval()`` (most of the time),
             but not a must. Please refer to the source code for more details.
         """
+        if ranks is None:
+            ranks = [1, 5, 10, 20]
         self.set_model_mode('eval')
         targets = list(self.test_loader.keys())
 
@@ -353,9 +357,11 @@ def _evaluate(
         visrank_topk=10,
         save_dir='',
         use_metric_cuhk03=False,
-        ranks=[1, 5, 10, 20],
+            ranks=None,
         rerank=False
     ):
+        if ranks is None:
+            ranks = [1, 5, 10, 20]
         batch_time = AverageMeter()
 
         def _feature_extraction(data_loader):

object_tracking/reid/torchreid/models/densenet.py

@@ -271,7 +271,7 @@ def init_pretrained_weights(model, model_url):
     # They are also in the checkpoints in model_urls. This pattern is used
     # to find such keys.
     pattern = re.compile(
-        r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$'
+        r'^(.*denselayer\d+\.(?:norm|relu|conv))\.([12]\.(?:weight|bias|running_mean|running_var))$'
     )
     for key in list(pretrain_dict.keys()):
         res = pattern.match(key)

object_tracking/reid/torchreid/models/hacnn.py

@@ -211,13 +211,15 @@ def __init__(
         self,
         num_classes,
         loss='softmax',
-        nchannels=[128, 256, 384],
+            nchannels=None,
         feat_dim=512,
         learn_region=True,
         use_gpu=True,
         **kwargs
     ):
         super(HACNN, self).__init__()
+        if nchannels is None:
+            nchannels = [128, 256, 384]
         self.loss = loss
         self.learn_region = learn_region
         self.use_gpu = use_gpu
@@ -400,7 +402,7 @@ def forward(self, x):
 
         if self.loss == 'softmax':
             if self.learn_region:
-                return (prelogits_global, prelogits_local)
+                return prelogits_global, prelogits_local
             else:
                 return prelogits_global
 

object_tracking/reid/torchreid/models/mlfn.py

@@ -111,11 +111,13 @@ def __init__(
         num_classes,
         loss='softmax',
         groups=32,
-        channels=[64, 256, 512, 1024, 2048],
+            channels=None,
         embed_dim=1024,
         **kwargs
     ):
         super(MLFN, self).__init__()
+        if channels is None:
+            channels = [64, 256, 512, 1024, 2048]
         self.loss = loss
         self.groups = groups
 

object_tracking/reid/torchreid/models/mobilenetv2.py

@@ -130,8 +130,7 @@ def _make_layer(self, block, t, c, n, s):
         # c: output channels
         # n: number of blocks
         # s: stride for first layer
-        layers = []
-        layers.append(block(self.in_channels, c, t, s))
+        layers = [block(self.in_channels, c, t, s)]
         self.in_channels = c
         for i in range(1, n):
             layers.append(block(self.in_channels, c, t))

object_tracking/reid/torchreid/models/osnet.py

@@ -351,9 +351,8 @@ def _make_layer(
         reduce_spatial_size,
         IN=False
     ):
-        layers = []
+        layers = [block(in_channels, out_channels, IN=IN)]
 
-        layers.append(block(in_channels, out_channels, IN=IN))
         for i in range(1, layer):
             layers.append(block(out_channels, out_channels, IN=IN))
 

object_tracking/reid/torchreid/models/pcb.py

@@ -109,13 +109,9 @@ class DimReduceLayer(nn.Module):
 
     def __init__(self, in_channels, out_channels, nonlinear):
         super(DimReduceLayer, self).__init__()
-        layers = []
-        layers.append(
-            nn.Conv2d(
-                in_channels, out_channels, 1, stride=1, padding=0, bias=False
-            )
-        )
-        layers.append(nn.BatchNorm2d(out_channels))
+        layers = [nn.Conv2d(
+            in_channels, out_channels, 1, stride=1, padding=0, bias=False
+        ), nn.BatchNorm2d(out_channels)]
 
         if nonlinear == 'relu':
             layers.append(nn.ReLU(inplace=True))
@@ -199,8 +195,7 @@ def _make_layer(self, block, planes, blocks, stride=1):
                 nn.BatchNorm2d(planes * block.expansion),
             )
 
-        layers = []
-        layers.append(block(self.inplanes, planes, stride, downsample))
+        layers = [block(self.inplanes, planes, stride, downsample)]
         self.inplanes = planes * block.expansion
         for i in range(1, blocks):
             layers.append(block(self.inplanes, planes))

object_tracking/reid/torchreid/models/resnet.py

@@ -267,13 +267,10 @@ def _make_layer(self, block, planes, blocks, stride=1, dilate=False):
                 norm_layer(planes * block.expansion),
             )
 
-        layers = []
-        layers.append(
-            block(
-                self.inplanes, planes, stride, downsample, self.groups,
-                self.base_width, previous_dilation, norm_layer
-            )
-        )
+        layers = [block(
+            self.inplanes, planes, stride, downsample, self.groups,
+            self.base_width, previous_dilation, norm_layer
+        )]
         self.inplanes = planes * block.expansion
         for _ in range(1, blocks):
             layers.append(

object_tracking/reid/torchreid/models/resnet_ibn_b.py

@@ -181,8 +181,7 @@ def _make_layer(self, block, planes, blocks, stride=1, IN=False):
                 nn.BatchNorm2d(planes * block.expansion),
             )
 
-        layers = []
-        layers.append(block(self.inplanes, planes, stride, downsample))
+        layers = [block(self.inplanes, planes, stride, downsample)]
         self.inplanes = planes * block.expansion
         for i in range(1, blocks - 1):
             layers.append(block(self.inplanes, planes))

object_tracking/reid/torchreid/models/resnetmid.py

@@ -169,8 +169,7 @@ def _make_layer(self, block, planes, blocks, stride=1):
                 nn.BatchNorm2d(planes * block.expansion),
             )
 
-        layers = []
-        layers.append(block(self.inplanes, planes, stride, downsample))
+        layers = [block(self.inplanes, planes, stride, downsample)]
         self.inplanes = planes * block.expansion
         for i in range(1, blocks):
             layers.append(block(self.inplanes, planes))

object_tracking/reid/torchreid/models/senet.py

@@ -438,12 +438,9 @@ def _make_layer(
                 nn.BatchNorm2d(planes * block.expansion),
             )
 
-        layers = []
-        layers.append(
-            block(
-                self.inplanes, planes, groups, reduction, stride, downsample
-            )
-        )
+        layers = [block(
+            self.inplanes, planes, groups, reduction, stride, downsample
+        )]
         self.inplanes = planes * block.expansion
         for i in range(1, blocks):
             layers.append(block(self.inplanes, planes, groups, reduction))

object_tracking/reid/torchreid/utils/feature_extractor.py

@@ -61,13 +61,17 @@ def __init__(
         model_name='',
         model_path='',
         image_size=(256, 128),
-        pixel_mean=[0.485, 0.456, 0.406],
-        pixel_std=[0.229, 0.224, 0.225],
+            pixel_mean=None,
+            pixel_std=None,
         pixel_norm=True,
         device='cuda',
         verbose=True
     ):
         # Build model
+        if pixel_mean is None:
+            pixel_mean = [0.485, 0.456, 0.406]
+        if pixel_std is None:
+            pixel_std = [0.229, 0.224, 0.225]
         model = build_model(
             model_name,
             num_classes=1,

object_tracking/test/deepsort_v3.py

@@ -93,8 +93,7 @@ def tracking(config: dict, json_save_dir: str, vis_save_dir: str, verbose=True):
                 id_num = str(track.track_id) #Get the ID for the particular track.
                 features = track.features #Get the feature vector corresponding to the detection.
 
-                track_dict = {}
-                track_dict["id"] = id_num
+                track_dict = {"id": id_num}
 
                 ans_box = get_closest_box(detections_class, bbox)
                 bbox = ans_box

object_tracking/test/deepsort_v4.py

@@ -89,8 +89,7 @@ def tracking(config: dict, json_save_dir: str, vis_save_dir: str, verbose=False)
                 bbox = track.to_tlbr() #Get the corrected/predicted bounding box
                 id_num = str(track.track_id) #Get the ID for the particular track.
                 feature = track.last_feature #Get the feature vector corresponding to the detection.
-                track_dict = {}
-                track_dict["id"] = id_num
+                track_dict = {"id": id_num}
 
                 ans_box = get_closest_box(detections_class, bbox)
                 bbox = ans_box

object_tracking/test/evaluate_subject.py

@@ -34,15 +34,15 @@ def evaluate(gt_boxes: list, cand_tracks: list):
             gt_box = gt_boxes[frame_idx]
             inside = is_box_in_box(gt_box, track_box)
 
-            if inside == True:
+            if inside:
                 iou = 1.0
             else:
                 iou = calculate_iou(gt_box, track_box)
 
             # if track.track_id == '400':
             dist = calculate_distance(gt_box, track_box)
 
-            if (iou > ACCEPT_IOU_THRES): #or (inside == True):
+            if iou > ACCEPT_IOU_THRES: #or (inside == True):
                 if start_frame == -1:
                     start_frame = frame_idx
                 end_frame = frame_idx

object_tracking/test/refine_v2.py

@@ -12,7 +12,7 @@
 
 def get_center_point(box: list):
     # xyxy
-    return ((box[0]+box[1])/2, (box[2]+box[3])/2)
+    return (box[0] + box[1]) / 2, (box[2] + box[3]) / 2
 
 def check_wrong_boxes(track_data: TrackResult):
     dist_meter = AverageMeter()