HongChuang/ultralytics-ascend
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 3

ultralytics 8.3.26 EdgeTPU Pose models fix (#17281)

Browse source 
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
This commit is contained in:
Laughing 2024-10-31 20:36:27 +08:00 • committed by GitHub
parent e8743f2ac9
commit f4e7756bff
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 19 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ultralytics/__init__.py
View file

@ -1,6 +1,6 @@
# Ultralytics YOLO 🚀, AGPL-3.0 license # Ultralytics YOLO 🚀, AGPL-3.0 license
__version__ = "8.3.25" __version__ = "8.3.26"
import os import os

3
ultralytics/nn/autobackend.py
View file

@ -663,6 +663,9 @@ class AutoBackend(nn.Module):
else: else:
x[:, [0, 2]] *= w x[:, [0, 2]] *= w
x[:, [1, 3]] *= h x[:, [1, 3]] *= h
if self.task == "pose":
x[:, 5::3] *= w
x[:, 6::3] *= h
y.append(x) y.append(x)
# TF segment fixes: export is reversed vs ONNX export and protos are transposed # TF segment fixes: export is reversed vs ONNX export and protos are transposed
if len(y) == 2: # segment with (det, proto) output order reversed if len(y) == 2: # segment with (det, proto) output order reversed

18
ultralytics/nn/modules/head.py
View file

@ -246,9 +246,21 @@ class Pose(Detect):
def kpts_decode(self, bs, kpts): def kpts_decode(self, bs, kpts):
"""Decodes keypoints.""" """Decodes keypoints."""
ndim = self.kpt_shape[1] ndim = self.kpt_shape[1]
if self.export: # required for TFLite export to avoid 'PLACEHOLDER_FOR_GREATER_OP_CODES' bug if self.export:
y = kpts.view(bs, *self.kpt_shape, -1) if self.format in {
a = (y[:, :, :2] * 2.0 + (self.anchors - 0.5)) * self.strides "tflite",
"edgetpu",
}: # required for TFLite export to avoid 'PLACEHOLDER_FOR_GREATER_OP_CODES' bug
# Precompute normalization factor to increase numerical stability
y = kpts.view(bs, *self.kpt_shape, -1)
grid_h, grid_w = self.shape[2], self.shape[3]
grid_size = torch.tensor([grid_w, grid_h], device=y.device).reshape(1, 2, 1)
norm = self.strides / (self.stride[0] * grid_size)
a = (y[:, :, :2] * 2.0 + (self.anchors - 0.5)) * norm
else:
# NCNN fix
y = kpts.view(bs, *self.kpt_shape, -1)
a = (y[:, :, :2] * 2.0 + (self.anchors - 0.5)) * self.strides
if ndim == 3: if ndim == 3:
a = torch.cat((a, y[:, :, 2:3].sigmoid()), 2) a = torch.cat((a, y[:, :, 2:3].sigmoid()), 2)
return a.view(bs, self.nk, -1) return a.view(bs, self.nk, -1)