End2end (#61)

* export end2end onnx model

* fixbug

* add web demo (#58)

* Update README.md

* main code

update yolov7-tiny deploy cfg

* main code

update yolov7-tiny training cfg

* main code

@liguagua752109150 #33 (comment)

* main code

@albertfaromatics #35 (comment)

* main code

update link

* main code

add custom hyp

* main code

update default activation function

* main code

update path

* main figure

add more tasks

* main code

update readme

* main code

update reparameterization

* Update README.md

* main code

update readme

* main code

update aux training

* main code

update aux training

* main code

update aux training

* main figure

update yolov7 prediction

* main code

update readme

* main code

rename

* main code

rename

* main code

rename

* main code

rename

* main code

update readme

* main code

update visualization

* main code

fix gain for train_aux

* main code

update loss

* main code

update instance segmentation demo

* main code

update keypoint detection demo

* main code

update pose demo

* main code

update pose

* main code

update pose

* main code

update pose

* main code

update pose

* main code

update trace

* Update README.md

* main code

fix ciou

* main code

fix nan of aux training #250 (comment) @hudingding

* support onnx to tensorrt convert (#114)

* fuse IDetect (#148)

* Fixes #199 (#203)

* minor fix

* resolve conflict

* resolve conflict

* resolve conflict

* resolve conflict

* resolve conflict

* resolve

* resolve

* resolve

* resolve

Co-authored-by: AK391 <[email protected]>
Co-authored-by: Alexey <[email protected]>
Co-authored-by: Kin-Yiu, Wong <[email protected]>
Co-authored-by: linghu8812 <[email protected]>
Co-authored-by: Alexander <[email protected]>
Co-authored-by: Ben Raymond <[email protected]>
Co-authored-by: AlexeyAB84 <[email protected]>

README.md

@@ -151,6 +151,13 @@ python detect.py --weights yolov7.pt --conf 0.25 --img-size 640 --source inferen
     </a>
 </div>
 
+
+## Export
+Use the args `--include-nms` can to export end to end onnx model which include the `EfficientNMS`. 
+```shell
+python models/export.py --weights yolov7.pt --grid  --include-nms
+```
+
 ## Citation
 
 ```

export.py

@@ -12,6 +12,7 @@
 from utils.activations import Hardswish, SiLU
 from utils.general import set_logging, check_img_size
 from utils.torch_utils import select_device
+from utils.add_nms import RegisterNMS
 
 if __name__ == '__main__':
     parser = argparse.ArgumentParser()
@@ -22,6 +23,7 @@
     parser.add_argument('--grid', action='store_true', help='export Detect() layer grid')
     parser.add_argument('--device', default='cpu', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
     parser.add_argument('--simplify', action='store_true', help='simplify onnx model')
+    parser.add_argument('--include-nms', action='store_true', help='export end2end onnx')
     opt = parser.parse_args()
     opt.img_size *= 2 if len(opt.img_size) == 1 else 1  # expand
     print(opt)
@@ -52,7 +54,9 @@
         #     m.forward = m.forward_export  # assign forward (optional)
     model.model[-1].export = not opt.grid  # set Detect() layer grid export
     y = model(img)  # dry run
-
+    if opt.include_nms:
+        model.model[-1].include_nms = True
+        y = None
     # TorchScript export
     try:
         print('\nStarting TorchScript export with torch %s...' % torch.__version__)
@@ -75,16 +79,23 @@
                           dynamic_axes={'images': {0: 'batch', 2: 'height', 3: 'width'},  # size(1,3,640,640)
                                         'output': {0: 'batch', 2: 'y', 3: 'x'}} if opt.dynamic else None)
 
-        # Checks
-        onnx_model = onnx.load(f)  # load onnx model
-        onnx.checker.check_model(onnx_model)  # check onnx model
-
-        # # Metadata
-        # d = {'stride': int(max(model.stride))}
-        # for k, v in d.items():
-        #     meta = onnx_model.metadata_props.add()
-        #     meta.key, meta.value = k, str(v)
-        # onnx.save(onnx_model, f)
+        if opt.include_nms:
+            print('Registering NMS plugin...')
+            mo = RegisterNMS(f)
+            mo.register_nms()
+            mo.save(f)
+        else:
+            # Checks
+            onnx_model = onnx.load(f)  # load onnx model
+            onnx.checker.check_model(onnx_model)  # check onnx model
+            # print(onnx.helper.printable_graph(onnx_model.graph))  # print a human readable model
+
+            # # Metadata
+            # d = {'stride': int(max(model.stride))}
+            # for k, v in d.items():
+            #     meta = onnx_model.metadata_props.add()
+            #     meta.key, meta.value = k, str(v)
+            # onnx.save(onnx_model, f)
 
         if opt.simplify:
             try:
@@ -95,11 +106,9 @@
                 assert check, 'assert check failed'
             except Exception as e:
                 print(f'Simplifier failure: {e}')
-        # print(onnx.helper.printable_graph(onnx_model.graph))  # print a human readable model
         print('ONNX export success, saved as %s' % f)
     except Exception as e:
         print('ONNX export failure: %s' % e)
-
     # CoreML export
     try:
         import coremltools as ct

models/common.py

@@ -236,7 +236,7 @@ def forward(self, x):
 class ResX(Res):
     # ResNet bottleneck
     def __init__(self, c1, c2, shortcut=True, g=32, e=0.5):  # ch_in, ch_out, shortcut, groups, expansion
-        super().__init__(c1, c2, shortcu, g, e)
+        super().__init__(c1, c2, shortcut, g, e)
         c_ = int(c2 * e)  # hidden channels
 
 

models/yolo.py

@@ -5,7 +5,7 @@
 
 sys.path.append('./')  # to run '$ python *.py' files in subdirectories
 logger = logging.getLogger(__name__)
-
+import torch
 from models.common import *
 from models.experimental import *
 from utils.autoanchor import check_anchor_order
@@ -23,7 +23,7 @@
 class Detect(nn.Module):
     stride = None  # strides computed during build
     export = False  # onnx export
-
+    include_nms = False 
     def __init__(self, nc=80, anchors=(), ch=()):  # detection layer
         super(Detect, self).__init__()
         self.nc = nc  # number of classes
@@ -48,7 +48,6 @@ def forward(self, x):
             if not self.training:  # inference
                 if self.grid[i].shape[2:4] != x[i].shape[2:4]:
                     self.grid[i] = self._make_grid(nx, ny).to(x[i].device)
-
                 y = x[i].sigmoid()
                 if not torch.onnx.is_in_onnx_export():
                     y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i]  # xy
@@ -59,13 +58,28 @@ def forward(self, x):
                     y = torch.cat((xy, wh, y[..., 4:]), -1)
                 z.append(y.view(bs, -1, self.no))
 
-        return x if self.training else (torch.cat(z, 1), x)
+        if self.include_nms:
+            z = self.convert(z)
+
+        return x if self.training else (z, ) if self.include_nms else (torch.cat(z, 1), x)
 
     @staticmethod
     def _make_grid(nx=20, ny=20):
         yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)])
         return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float()
 
+    def convert(self, z):
+        z = torch.cat(z, 1)
+        box = z[:, :, :4]
+        conf = z[:, :, 4:5]
+        score = z[:, :, 5:]
+        score *= conf
+        convert_matrix = torch.tensor([[1, 0, 1, 0], [0, 1, 0, 1], [-0.5, 0, 0.5, 0], [0, -0.5, 0, 0.5]],
+                                           dtype=torch.float32,
+                                           device=z.device)
+        box @= convert_matrix                          
+        return (box, score)
+
 
 class IDetect(nn.Module):
     stride = None  # strides computed during build