-
```
run non-zero-copy model
Read model/test1.jpg ...
img width = 640, img height = 480
Loading mode...
[ 1] HAL user version 6.4.3.5.293908
[ 2] HAL kernel version 6.4.3.5.293908
./run_…
-
========== parser_config ==========
NPU_VERSION:1
RK_device_platform:RK1808
input_example:['/workspace/data/BackgroundIdentification/data_0331_1000/images/train/20210706141643_1_19.jpg']
pre_compi…
-
There's 'Not support input data type 'float16'!' during onnx model conversion.
The target platform is rk3588.
The working environment is Ubuntu 20.04.5, Python 3.8.10, rknn-toolkit2 1.4.0.
Onnx m…
-
Hello! Thank you for your impressive work.
I’m trying to run RT-DETR on RK3588, but I’m having trouble converting the model to rknn format.
The exported .onnx model requires opset>=16, but rknn on…
-
dataset一张图片没问题,图片多久出错
done
--> Building model
W:tensorflow:From /home/xyz/miniconda3/envs/rknn/lib/python3.6/site-packages/rknn/api/rknn.py:257: The name tf.reset_default_graph is deprecated. Ple…
-
hey! i am working with yolov8pose-onnx-rknn. i don't have any experience before. Can you give me some sample as your YOLOv8--Segmentation when you have time to do.Thanks a lot!!!
-
D Real output shape: (2, 4, 4, 1792)
I Build torch-jit-export complete.
I Clean.
MMSE Quant Step 0: 96%|████████████████████████████████████████▏ | 316/330 [00:30
-
请问为什么在使用torch.argmax()时,转rknn报出如下错误:
E Catch exception when building RKNN model!
E Traceback (most recent call last):
E File "rknn/api/rknn_base.py", line 616, in rknn.api.rknn_base.RKNNBase.…
-
-
我使用RKNNLite库去推理一个视频流, 进行一次rknn.init_runtime 然后不断的推理,发现内存会一直上涨直到崩溃。
后来我改为每一帧都rknn.init_runtime, 推理完这一帧后release, 内存问题泄漏就没有了,
就是应该这样做吗?