Merge remote-tracking branch 'upstream/master'

detect.py

@@ -18,7 +18,7 @@ def detect(save_img=False):
 
     # Load model
     google_utils.attempt_download(weights)
-    model = torch.load(weights, map_location=device)['model']
+    model = torch.load(weights, map_location=device)['model'].float()  # load to FP32
     # torch.save(torch.load(weights, map_location=device), weights)  # update model if SourceChangeWarning
     # model.fuse()
     model.to(device).eval()

hubconf.py

@@ -32,8 +32,8 @@ def create(name, pretrained, channels, classes):
     if pretrained:
         ckpt = '%s.pt' % name  # checkpoint filename
         google_utils.attempt_download(ckpt)  # download if not found locally
-        state_dict = torch.load(ckpt, map_location=torch.device('cpu'))['model'].state_dict()
+        state_dict = torch.load(ckpt, map_location=torch.device('cpu'))['model'].float().state_dict()  # to FP32
-        state_dict = {k: v for k, v in state_dict.items() if model.state_dict()[k].numel() == v.numel()}  # filter
+        state_dict = {k: v for k, v in state_dict.items() if model.state_dict()[k].shape == v.shape}  # filter
         model.load_state_dict(state_dict, strict=False)  # load
     return model
 

test.py

@@ -23,6 +23,7 @@ def test(data,
          verbose=False):
     # Initialize/load model and set device
     if model is None:
+        training = False
         device = torch_utils.select_device(opt.device, batch_size=batch_size)
         half = device.type != 'cpu'  # half precision only supported on CUDA
 
@@ -32,9 +33,9 @@ def test(data,
 
         # Load model
         google_utils.attempt_download(weights)
-        model = torch.load(weights, map_location=device)['model']
+        model = torch.load(weights, map_location=device)['model'].float()  # load to FP32
         torch_utils.model_info(model)
-        # model.fuse()
+        model.fuse()
         model.to(device)
         if half:
             model.half()  # to FP16
@@ -42,11 +43,12 @@ def test(data,
         if device.type != 'cpu' and torch.cuda.device_count() > 1:
             model = nn.DataParallel(model)
 
-        training = False
     else:  # called by train.py
-        device = next(model.parameters()).device  # get model device
-        half = False
         training = True
+        device = next(model.parameters()).device  # get model device
+        half = device.type != 'cpu'  # half precision only supported on CUDA
+        if half:
+            model.half()  # to FP16
 
     # Configure
     model.eval()
@@ -69,7 +71,7 @@ def test(data,
                                       batch_size,
                                       rect=True,  # rectangular inference
                                       single_cls=opt.single_cls,  # single class mode
-                                      pad=0.0 if fast else 0.5)  # padding
+                                      pad=0.5)  # padding
         batch_size = min(batch_size, len(dataset))
         nw = min([os.cpu_count(), batch_size if batch_size > 1 else 0, 8])  # number of workers
         dataloader = DataLoader(dataset,
@@ -102,7 +104,7 @@ def test(data,
 
             # Compute loss
             if training:  # if model has loss hyperparameters
-                loss += compute_loss(train_out, targets, model)[1][:3]  # GIoU, obj, cls
+                loss += compute_loss([x.float() for x in train_out], targets, model)[1][:3]  # GIoU, obj, cls
 
             # Run NMS
             t = torch_utils.time_synchronized()
@@ -255,7 +257,7 @@ if __name__ == '__main__':
     opt = parser.parse_args()
     opt.img_size = check_img_size(opt.img_size)
     opt.save_json = opt.save_json or opt.data.endswith('coco.yaml')
-    opt.data = glob.glob('./**/' + opt.data, recursive=True)[0]  # find file
+    opt.data = check_file(opt.data)  # check file
     print(opt)
 
     # task = 'val', 'test', 'study'

train.py

@@ -112,8 +112,8 @@ def train(hyp):
 
         # load model
         try:
-            ckpt['model'] = \
+            ckpt['model'] = {k: v for k, v in ckpt['model'].float().state_dict().items()
-                {k: v for k, v in ckpt['model'].state_dict().items() if model.state_dict()[k].numel() == v.numel()}
+                             if model.state_dict()[k].shape == v.shape}  # to FP32, filter
             model.load_state_dict(ckpt['model'], strict=False)
         except KeyError as e:
             s = "%s is not compatible with %s. Specify --weights '' or specify a --cfg compatible with %s." \
@@ -363,6 +363,7 @@ def train(hyp):
 
 
 if __name__ == '__main__':
+    check_git_status()
     parser = argparse.ArgumentParser()
     parser.add_argument('--epochs', type=int, default=300)
     parser.add_argument('--batch-size', type=int, default=16)
@@ -384,12 +385,11 @@ if __name__ == '__main__':
     parser.add_argument('--single-cls', action='store_true', help='train as single-class dataset')
     opt = parser.parse_args()
     opt.weights = last if opt.resume else opt.weights
-    opt.cfg = glob.glob('./**/' + opt.cfg, recursive=True)[0]  # find file
+    opt.cfg = check_file(opt.cfg)  # check file
-    opt.data = glob.glob('./**/' + opt.data, recursive=True)[0]  # find file
+    opt.data = check_file(opt.data)  # check file
     print(opt)
     opt.img_size.extend([opt.img_size[-1]] * (2 - len(opt.img_size)))  # extend to 2 sizes (train, test)
     device = torch_utils.select_device(opt.device, apex=mixed_precision, batch_size=opt.batch_size)
-    # check_git_status()
     if device.type == 'cpu':
         mixed_precision = False
 

utils/activations.py

@@ -1,4 +1,5 @@
 import torch
+import torch.nn as nn
 import torch.nn.functional as F
 import torch.nn as nn
 

utils/google_utils.py

@@ -25,9 +25,14 @@ def attempt_download(weights):
         if file in d:
             r = gdrive_download(id=d[file], name=weights)
 
+        if not (r == 0 and os.path.exists(weights) and os.path.getsize(weights) > 1E6):  # weights exist and > 1MB
+            os.remove(weights) if os.path.exists(weights) else None  # remove partial downloads
+            s = "curl -L -o %s 'https://storage.googleapis.com/ultralytics/yolov5/ckpt/%s'" % (weights, file)
+            r = os.system(s)  # execute, capture return values
+
             # Error check
             if not (r == 0 and os.path.exists(weights) and os.path.getsize(weights) > 1E6):  # weights exist and > 1MB
-            os.system('rm ' + weights)  # remove partial downloads
+                os.remove(weights) if os.path.exists(weights) else None  # remove partial downloads
                 raise Exception(msg)
 
 

utils/utils.py

@@ -64,6 +64,16 @@ def check_best_possible_recall(dataset, anchors, thr):
                       'Compute new anchors with utils.utils.kmeans_anchors() and update model before training.' % bpr
 
 
+def check_file(file):
+    # Searches for file if not found locally
+    if os.path.isfile(file):
+        return file
+    else:
+        files = glob.glob('./**/' + file, recursive=True)  # find file
+        assert len(files), 'File Not Found: %s' % file  # assert file was found
+        return files[0]  # return first file if multiple found
+
+
 def make_divisible(x, divisor):
     # Returns x evenly divisble by divisor
     return math.ceil(x / divisor) * divisor
@@ -518,7 +528,7 @@ def non_max_suppression(prediction, conf_thres=0.1, iou_thres=0.6, fast=False, c
     fast |= conf_thres > 0.001  # fast mode
     if fast:
         merge = False
-        multi_label = False
+        multi_label = nc > 1  # multiple labels per box (adds 0.5ms/img)
     else:
         merge = True  # merge for best mAP (adds 0.5ms/img)
         multi_label = nc > 1  # multiple labels per box (adds 0.5ms/img)