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

ultralytics 8.3.72 Fix NVIDIA Jetson DLA core support for DLA inference (#19078)

Browse source 
Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
Co-authored-by: Lakshantha Dissanayake <lakshanthad@yahoo.com>
Co-authored-by: Lakshantha Dissanayake <lakshantha@ultralytics.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
This commit is contained in:
Laughing 2025-02-06 11:10:36 +08:00 • committed by GitHub
parent 84a8b067c4
commit c1860b8333
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
5 changed files with 15 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

2
.github/workflows/ci.yml vendored
View file

@ -99,7 +99,7 @@ jobs:
fail-fast: false
matrix:
# Temporarily disable windows-latest due to https://github.com/ultralytics/ultralytics/actions/runs/13020330819/job/36319338854?pr=18921
os: [ubuntu-latest, macos-15, ubuntu-24.04-arm]
os: [ubuntu-latest, macos-15]
python-version: ["3.11"]
model: [yolo11n]
steps:

12
docs/en/guides/nvidia-jetson.md
View file

@ -289,10 +289,13 @@ The YOLO11n model in PyTorch format is converted to TensorRT to run inference wi
The following Jetson devices are equipped with DLA hardware:
- Jetson Orin NX 16GB
- Jetson AGX Orin Series
- Jetson AGX Xavier Series
- Jetson Xavier NX Series
| Jetson Device | DLA Cores | DLA Max Frequency |
| ------------------------ | --------- | ----------------- |
| Jetson AGX Orin Series | 2 | 1.6 GHz |
| Jetson Orin NX 16GB | 2 | 614 MHz |
| Jetson Orin NX 8GB | 1 | 614 MHz |
| Jetson AGX Xavier Series | 2 | 1.4 GHz |
| Jetson Xavier NX Series | 2 | 1.1 GHz |
!!! example
@ -318,6 +321,7 @@ The following Jetson devices are equipped with DLA hardware:
```bash
# Export a YOLO11n PyTorch model to TensorRT format with DLA enabled (only works with FP16 or INT8)
# Once DLA core number is specified at export, it will use the same core at inference
yolo export model=yolo11n.pt format=engine device="dla:0" half=True # dla:0 or dla:1 corresponds to the DLA cores
# Run inference with the exported model on the DLA

2
ultralytics/__init__.py
View file

@ -1,6 +1,6 @@
# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
__version__ = "8.3.71"
__version__ = "8.3.72"
import os

2
ultralytics/engine/exporter.py
View file

@ -386,6 +386,8 @@ class Exporter:
"names": model.names,
"args": {k: v for k, v in self.args if k in fmt_keys},
} # model metadata
if dla is not None:
self.metadata["dla"] = dla # make sure `AutoBackend` uses correct dla device if it has one
if model.task == "pose":
self.metadata["kpt_shape"] = model.model[-1].kpt_shape

9
ultralytics/nn/autobackend.py
View file

@ -292,13 +292,10 @@ class AutoBackend(nn.Module):
metadata = json.loads(f.read(meta_len).decode("utf-8")) # read metadata
except UnicodeDecodeError:
f.seek(0) # engine file may lack embedded Ultralytics metadata
dla = metadata.get("dla", None)
if dla is not None:
runtime.DLA_core = int(dla)
model = runtime.deserialize_cuda_engine(f.read()) # read engine
if "dla" in str(device.type):
dla_core = int(device.type.split(":")[1])
assert dla_core in {0, 1}, (
"Expected device type for inference in DLA is 'dla:0' or 'dla:1', but received '{device.type}'"
)
runtime.DLA_core = dla_core
# Model context
try: