rknpu2
rknpu2 copied to clipboard
rockchip-linux
E RKNN: [08:24:12.287] failed to submit!
我们这个模型转换成功了，但不能跑是什么情况？错误如下：
sdk version: 1.5.0 (e6fe0c678@2023-05-25T08:09:20) driver version: 0.8.2
model input num: 1, output num: 1
index=0, name=input, n_dims=4, dims=[1, 224, 224, 3], n_elems=150528, size=150528, w_stride = 224, size_with_stride=150528, fmt=NHWC, type=INT8, qnt_type=AFFINE, zp=-128, scale=1.000000
index=0, name=1386, n_dims=2, dims=[1, 6], n_elems=6, size=6, w_stride = 0, size_with_stride=6, fmt=UNDEFINED, type=INT8, qnt_type=AFFINE, zp=-128, scale=0.109407
model is NHWC input fmt
model input height=224, width=224, channel=3
Start rknn_inputs_set!!
rknn_inputs_set DONE!!
E RKNN: [08:24:12.287] failed to submit!, op id: 1, op name: Conv:Conv_0_stride_split, flags: 0x5, task start: 0, task number: 18, run task counter: 0, int status: 0, please try updating to the latest version of the toolkit2 and runtime from: https://eyun.baidu.com/s/3eTDMk6Y (PWD: rknn)
i think u should make it to one input and three output
It's not a yolo-like model, why I need to divide my output into three?
你好，你的问题解决了吗？我遇到了类似的问题，我将语音识别模型转成rknn之后，放到rk3588板子上推理的时候报错： E RKNN: [10:12:36.185] failed to submit!, op id: 51, op name: Conv:/wav2vec2/encoder/pos_conv_embed/conv/Conv_pads_conv_group_0, flags: 0x5, task start: 3177, task number: 353, run task counter: 288, int status: 0, please try updating to the latest version of the toolkit2 and runtime from: https://eyun.baidu.com/s/3eTDMk6Y (PWD: rknn)
但是，我在模拟器上推理的时候，是能够有正常的识别结果的
你好，你的问题解决了吗？我遇到了类似的问题，我将语音识别模型转成rknn之后，放到rk3588板子上推理的时候报错： E RKNN: [10:12:36.185] failed to submit!, op id: 51, op name: Conv:/wav2vec2/encoder/pos_conv_embed/conv/Conv_pads_conv_group_0, flags: 0x5, task start: 3177, task number: 353, run task counter: 288, int status: 0, please try updating to the latest version of the toolkit2 and runtime from: https://eyun.baidu.com/s/3eTDMk6Y (PWD: rknn)
但是，我在模拟器上推理的时候，是能够有正常的识别结果的
你解决了吗，我在跑yolov9的时候也遇到了这个问题
你好，你的问题解决了吗？我遇到了类似的问题，我将语音识别模型转成rknn之后，放到rk3588板子上推理的时候报错： E RKNN: [10:12:36.185] failed to submit!, op id: 51, op name: Conv:/wav2vec2/encoder/pos_conv_embed/conv/Conv_pads_conv_group_0, flags: 0x5, task start: 3177, task number: 353, run task counter: 288, int status: 0, please try updating to the latest version of the toolkit2 and runtime from: https://eyun.baidu.com/s/3eTDMk6Y (PWD: rknn) 但是，我在模拟器上推理的时候，是能够有正常的识别结果的
你解决了吗，我在跑yolov9的时候也遇到了这个问题
太久了 但是已经解决了，我记得好像是版本的一些问题吧，反正安装的版本要对应上。我当时装的时候toolkit和toolkit_lite都是1.5.0版本，还有主要你看看你板子上/usr/lib/librknnrt.so文件的版本，我当时是1.4的，和我装的toolkit_lite版本不太匹配，我就手动换成了1.5.2版本的了，就解决了
i think u should make it to one input and three output
It's not a yolo-like model, why I need to divide my output into three?
I made a mistake,sor…