thebears/deepstream_yolo
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

New features and fixes

Browse Source
This commit is contained in:
Marcos Luciano
2023-06-05 14:48:23 -03:00
parent 3f14b0d95d
commit 66a6754b77
57 changed files with 2137 additions and 1534 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
docs/YOLONAS.md
View File

@@ -46,6 +46,24 @@ Generate the ONNX model file (example for YOLO-NAS S)
python3 export_yolonas.py -m yolo_nas_s -w yolo_nas_s_coco.pth --simplify --dynamic
```
**NOTE**: To simplify the ONNX model
```
--simplify
```
**NOTE**: To use dynamic batch-size
```
--dynamic
```
**NOTE**: To use implicit batch-size (example for batch-size = 4)
```
--batch 4
```
**NOTE**: If you are using DeepStream 5.1, use opset 12 or lower. The default opset is 14.
```
@@ -128,7 +146,7 @@ Open the `DeepStream-Yolo` folder and compile the lib
* DeepStream 5.1 on x86 platform
```
CUDA_VER=11.1 LEGACY=1 make -C nvdsinfer_custom_impl_Yolo
CUDA_VER=11.1 make -C nvdsinfer_custom_impl_Yolo
```
* DeepStream 6.2 / 6.1.1 / 6.1 on Jetson platform
@@ -137,18 +155,12 @@ Open the `DeepStream-Yolo` folder and compile the lib
CUDA_VER=11.4 make -C nvdsinfer_custom_impl_Yolo
```
* DeepStream 6.0.1 / 6.0 on Jetson platform
* DeepStream 6.0.1 / 6.0 / 5.1 on Jetson platform
```
CUDA_VER=10.2 make -C nvdsinfer_custom_impl_Yolo
```
* DeepStream 5.1 on Jetson platform
```
CUDA_VER=10.2 LEGACY=1 make -C nvdsinfer_custom_impl_Yolo
```
##
### Edit the config_infer_primary_yolonas file
@@ -159,7 +171,6 @@ Edit the `config_infer_primary_yolonas.txt` file according to your model (exampl
[property]
...
onnx-file=yolo_nas_s_coco.onnx
model-engine-file=yolo_nas_s_coco.onnx_b1_gpu0_fp32.engine
...
num-detected-classes=80
...
@@ -170,8 +181,18 @@ parse-bbox-func-name=NvDsInferParseYoloE
**NOTE**: The **YOLO-NAS** resizes the input with left/top padding. To get better accuracy, use
```
...
maintain-aspect-ratio=1
symmetric-padding=0
...
```
**NOTE**: By default, the dynamic batch-size is set. To use implicit batch-size, uncomment the line
```
...
force-implicit-batch-dim=1
...
```
##