-
Notifications
You must be signed in to change notification settings - Fork 90
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Yolo v8 compile error #163
Comments
Using rknnlite inference, and then reporting errors? |
I gave an example of an error from From rknnlite inference on RKNN_MODEL = "yolov8_small.rknn"
rknn = RKNNLite()
ret = rknn.load_rknn(RKNN_MODEL)
ret = rknn.init_runtime()
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
input_frame = cv2.resize(frame, (640, 640))
input_frame = numpy.expand_dims(input_frame, 0)
outputs = rknn.inference(inputs=[input_frame])
print(outputs)
print(len(outputs))
print(type(outputs))
rknn.release() Result: W rknn-toolkit-lite2 version: 2.2.0
-> Start Program <-
--> Load RKNN model
done
--> Init runtime environment
I RKNN: [11:16:40.100] RKNN Runtime Information, librknnrt version: 2.2.0 (c195366594@2024-09-14T12:18:56)
I RKNN: [11:16:40.100] RKNN Driver Information, version: 0.9.6
I RKNN: [11:16:40.100] RKNN Model Information, version: 6, toolkit version: 2.2.0(compiler version: 2.2.0 (c195366594@2024-09-14T12:24:14)), target: RKNPU v2, target platform: rk3588, framework name: ONNX, framework layout: NCHW, model inference type: static_shape
W RKNN: [11:16:40.121] query RKNN_QUERY_INPUT_DYNAMIC_RANGE error, rknn model is static shape type, please export rknn with dynamic_shapes
W Query dynamic range failed. Ret code: RKNN_ERR_MODEL_INVALID. (If it is a static shape RKNN model, please ignore the above warning message.)
done
-> Loading Done <-
[array([[[ 0. , 0. , 0. , ..., 0. ,
0. , 0. ],
[ 0. , 0. , 0. , ..., 0. ,
0. , 0. ],
[ 0. , 0. , 0. , ..., 0. ,
0. , 0. ],
...,
[ 0. , 0. , 0. , ..., 0. ,
0. , 0. ],
[ 0. , 0. , 0. , ..., 0. ,
0. , 0. ],
[ 0. , 0. , 0. , ..., 0. ,
0. , 0. ]]], dtype=float32)]
1
<class 'list'> I think that the compilation of the model is incorrect, because when I run the inference of the original model, I get the following values for the test frame: [array([[[6.4593658e+00, 7.8225298e+00, 1.3197342e+01, ...,
3.5425361e-07, 1.6246098e-07, 1.0452623e-07],
[1.6069817e+01, 6.6863852e+00, 3.1918978e+01, ...,
2.1877072e-07, 1.0333743e-07, 8.5050246e-08],
[1.8130138e+01, 3.7029805e+00, 3.4330154e+01, ...,
3.2787889e-07, 8.8689447e-08, 1.2997523e-07],
...,
[5.3248120e+02, 6.0730432e+02, 2.3935321e+02, ...,
1.0074392e-06, 1.5418524e-06, 1.7887144e-06],
[5.6285797e+02, 6.0808728e+02, 2.1686890e+02, ...,
7.3178660e-07, 1.3735065e-06, 1.6389531e-06],
[6.0336908e+02, 6.0807568e+02, 9.7625366e+01, ...,
1.2430581e-06, 1.5659134e-06, 1.8450241e-06]]], dtype=float32)] |
Mistakenly closed the issue😅 |
It looks like an internal bug, which is not easy to solve. You can try to convert the model using a lower version of toolkit. |
--> Configuring model yolov8_small.onnx
done
--> Loading model
I Loading : 100%|█████████████████████████████████████████████| 143/143 [00:00<00:00, 195115.64it/s]
done
--> Building model
D base_optimize ...
D base_optimize done.
D
D fold_constant ...
D fold_constant done.
D
D correct_ops ...
D correct_ops done.
D
D fuse_ops ...
D fuse_ops results:
D replace_exswish: remove node = ['/model.0/act/Sigmoid', '/model.0/act/Mul'], add node = ['/model.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.1/act/Sigmoid', '/model.1/act/Mul'], add node = ['/model.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/cv1/act/Sigmoid', '/model.2/cv1/act/Mul'], add node = ['/model.2/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/m.0/cv1/act/Sigmoid', '/model.2/m.0/cv1/act/Mul'], add node = ['/model.2/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/m.0/cv2/act/Sigmoid', '/model.2/m.0/cv2/act/Mul'], add node = ['/model.2/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/cv2/act/Sigmoid', '/model.2/cv2/act/Mul'], add node = ['/model.2/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.3/act/Sigmoid', '/model.3/act/Mul'], add node = ['/model.3/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/cv1/act/Sigmoid', '/model.4/cv1/act/Mul'], add node = ['/model.4/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.0/cv1/act/Sigmoid', '/model.4/m.0/cv1/act/Mul'], add node = ['/model.4/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.0/cv2/act/Sigmoid', '/model.4/m.0/cv2/act/Mul'], add node = ['/model.4/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.1/cv1/act/Sigmoid', '/model.4/m.1/cv1/act/Mul'], add node = ['/model.4/m.1/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.1/cv2/act/Sigmoid', '/model.4/m.1/cv2/act/Mul'], add node = ['/model.4/m.1/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/cv2/act/Sigmoid', '/model.4/cv2/act/Mul'], add node = ['/model.4/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.5/act/Sigmoid', '/model.5/act/Mul'], add node = ['/model.5/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/cv1/act/Sigmoid', '/model.6/cv1/act/Mul'], add node = ['/model.6/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.0/cv1/act/Sigmoid', '/model.6/m.0/cv1/act/Mul'], add node = ['/model.6/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.0/cv2/act/Sigmoid', '/model.6/m.0/cv2/act/Mul'], add node = ['/model.6/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.1/cv1/act/Sigmoid', '/model.6/m.1/cv1/act/Mul'], add node = ['/model.6/m.1/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.1/cv2/act/Sigmoid', '/model.6/m.1/cv2/act/Mul'], add node = ['/model.6/m.1/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/cv2/act/Sigmoid', '/model.6/cv2/act/Mul'], add node = ['/model.6/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.7/act/Sigmoid', '/model.7/act/Mul'], add node = ['/model.7/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/cv1/act/Sigmoid', '/model.8/cv1/act/Mul'], add node = ['/model.8/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/m.0/cv1/act/Sigmoid', '/model.8/m.0/cv1/act/Mul'], add node = ['/model.8/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/m.0/cv2/act/Sigmoid', '/model.8/m.0/cv2/act/Mul'], add node = ['/model.8/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/cv2/act/Sigmoid', '/model.8/cv2/act/Mul'], add node = ['/model.8/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.9/cv1/act/Sigmoid', '/model.9/cv1/act/Mul'], add node = ['/model.9/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.9/cv2/act/Sigmoid', '/model.9/cv2/act/Mul'], add node = ['/model.9/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/cv1/act/Sigmoid', '/model.12/cv1/act/Mul'], add node = ['/model.12/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/m.0/cv1/act/Sigmoid', '/model.12/m.0/cv1/act/Mul'], add node = ['/model.12/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/m.0/cv2/act/Sigmoid', '/model.12/m.0/cv2/act/Mul'], add node = ['/model.12/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/cv2/act/Sigmoid', '/model.12/cv2/act/Mul'], add node = ['/model.12/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/cv1/act/Sigmoid', '/model.15/cv1/act/Mul'], add node = ['/model.15/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/m.0/cv1/act/Sigmoid', '/model.15/m.0/cv1/act/Mul'], add node = ['/model.15/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/m.0/cv2/act/Sigmoid', '/model.15/m.0/cv2/act/Mul'], add node = ['/model.15/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/cv2/act/Sigmoid', '/model.15/cv2/act/Mul'], add node = ['/model.15/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.16/act/Sigmoid', '/model.16/act/Mul'], add node = ['/model.16/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/cv1/act/Sigmoid', '/model.18/cv1/act/Mul'], add node = ['/model.18/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/m.0/cv1/act/Sigmoid', '/model.18/m.0/cv1/act/Mul'], add node = ['/model.18/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/m.0/cv2/act/Sigmoid', '/model.18/m.0/cv2/act/Mul'], add node = ['/model.18/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/cv2/act/Sigmoid', '/model.18/cv2/act/Mul'], add node = ['/model.18/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.19/act/Sigmoid', '/model.19/act/Mul'], add node = ['/model.19/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/cv1/act/Sigmoid', '/model.21/cv1/act/Mul'], add node = ['/model.21/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/m.0/cv1/act/Sigmoid', '/model.21/m.0/cv1/act/Mul'], add node = ['/model.21/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/m.0/cv2/act/Sigmoid', '/model.21/m.0/cv2/act/Mul'], add node = ['/model.21/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/cv2/act/Sigmoid', '/model.21/cv2/act/Mul'], add node = ['/model.21/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.2/cv2.2.0/act/Sigmoid', '/model.22/cv2.2/cv2.2.0/act/Mul'], add node = ['/model.22/cv2.2/cv2.2.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.2/cv2.2.1/act/Sigmoid', '/model.22/cv2.2/cv2.2.1/act/Mul'], add node = ['/model.22/cv2.2/cv2.2.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.2/cv3.2.0/act/Sigmoid', '/model.22/cv3.2/cv3.2.0/act/Mul'], add node = ['/model.22/cv3.2/cv3.2.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.2/cv3.2.1/act/Sigmoid', '/model.22/cv3.2/cv3.2.1/act/Mul'], add node = ['/model.22/cv3.2/cv3.2.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.1/cv2.1.0/act/Sigmoid', '/model.22/cv2.1/cv2.1.0/act/Mul'], add node = ['/model.22/cv2.1/cv2.1.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.1/cv2.1.1/act/Sigmoid', '/model.22/cv2.1/cv2.1.1/act/Mul'], add node = ['/model.22/cv2.1/cv2.1.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.1/cv3.1.0/act/Sigmoid', '/model.22/cv3.1/cv3.1.0/act/Mul'], add node = ['/model.22/cv3.1/cv3.1.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.1/cv3.1.1/act/Sigmoid', '/model.22/cv3.1/cv3.1.1/act/Mul'], add node = ['/model.22/cv3.1/cv3.1.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.0/cv2.0.0/act/Sigmoid', '/model.22/cv2.0/cv2.0.0/act/Mul'], add node = ['/model.22/cv2.0/cv2.0.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.0/cv2.0.1/act/Sigmoid', '/model.22/cv2.0/cv2.0.1/act/Mul'], add node = ['/model.22/cv2.0/cv2.0.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.0/cv3.0.0/act/Sigmoid', '/model.22/cv3.0/cv3.0.0/act/Mul'], add node = ['/model.22/cv3.0/cv3.0.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.0/cv3.0.1/act/Sigmoid', '/model.22/cv3.0/cv3.0.1/act/Mul'], add node = ['/model.22/cv3.0/cv3.0.1/act/Sigmoid_2swish']
D replace_parallel_slice_by_split: remove node = ['/model.22/Slice', '/model.22/Slice_1'], add node = ['/model.22/Slice_2sp']
D unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Reshape_output_0_rs', '/model.22/Reshape_1_output_0_rs', '/model.22/Reshape_2_output_0_rs', '/model.22/Concat_3_output_0-rs']
D unsqueeze_to_4d_split: remove node = [], add node = ['/model.22/Concat_3_output_0_rs', '/model.22/Split_output_0-rs', '/model.22/Split_output_1-rs']
D convert_softmax_to_exsoftmax13: remove node = ['/model.22/dfl/Softmax'], add node = ['/model.22/dfl/Softmax']
D unsqueeze_to_4d_split: remove node = [], add node = ['/model.22/dfl/Reshape_1_output_0_rs', '/model.22/Slice_output_0-rs', '/model.22/Slice_1_output_0-rs']
D unsqueeze_to_4d_sub: remove node = [], add node = ['/model.22/Slice_output_0_rs', '/model.22/Sub_output_0-rs']
D unsqueeze_to_4d_add: remove node = [], add node = ['/model.22/Slice_1_output_0_rs', '/model.22/Add_1_output_0-rs']
D unsqueeze_to_4d_add: remove node = [], add node = ['/model.22/Sub_output_0_rs', '/model.22/Add_1_output_0_rs', '/model.22/Add_2_output_0-rs']
D convert_div_to_mul: remove node = ['/model.22/Div_1'], add node = ['/model.22/Div_1_2mul']
D unsqueeze_to_4d_sub: remove node = [], add node = ['/model.22/Add_1_output_0_rs#1', '/model.22/Sub_output_0_rs#1', '/model.22/Sub_1_output_0-rs']
D unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Div_1_output_0_rs', '/model.22/Sub_1_output_0_rs', '/model.22/Concat_4_output_0-rs']
D unsqueeze_to_4d_mul: remove node = [], add node = ['/model.22/Concat_4_output_0_rs', '/model.22/Mul_2_output_0-rs']
D unsqueeze_to_4d_sigmoid: remove node = [], add node = ['/model.22/Split_output_1_rs', '/model.22/Sigmoid_output_0-rs']
D unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Mul_2_output_0_rs', '/model.22/Sigmoid_output_0_rs', 'output0-rs']
D unsqueeze_to_4d_mul: remove node = [], add node = ['/model.22/Add_2_output_0_rs', '/model.22/Div_1_output_0-rs']
D swap_concat_axis_avoid_channel_concat: remove node = [], add node = ['/model.22/Div_1_output_0_rs_rs', '/model.22/Sub_1_output_0_rs_rs', '/model.22/Concat_4_output_0-rs-rs']
D input_align_4D_mul: remove node = ['/model.22/Mul_2'], add node = ['/model.22/Mul_2']
D swap_concat_axis_avoid_channel_concat: remove node = [], add node = ['/model.22/Mul_2_output_0_rs_rs', '/model.22/Sigmoid_output_0_rs_rs', 'output0-rs-rs']
D convert_reshape_to_transpose: remove node = ['/model.22/Sub_1_output_0_rs_rs'], add node = ['/model.22/Sub_1_output_0_rs_rs']
D input_align_4D_mul: remove node = ['/model.22/Div_1_2mul'], add node = ['/model.22/Div_1_2mul']
D convert_reshape_to_transpose: remove node = ['/model.22/Div_1_output_0_rs_rs'], add node = ['/model.22/Div_1_output_0_rs_rs']
D convert_reshape_to_transpose: remove node = ['/model.22/Concat_4_output_0-rs-rs'], add node = ['/model.22/Concat_4_output_0-rs-rs']
D convert_reshape_to_transpose: remove node = ['/model.22/Mul_2_output_0_rs_rs'], add node = ['/model.22/Mul_2_output_0_rs_rs']
D convert_reshape_to_transpose: remove node = ['/model.22/Sigmoid_output_0_rs_rs'], add node = ['/model.22/Sigmoid_output_0_rs_rs']
D convert_reshape_to_transpose: remove node = ['output0-rs-rs'], add node = ['output0-rs-rs']
D fuse_two_reshape: remove node = ['/model.22/Reshape_2', '/model.22/Reshape_1', '/model.22/Reshape']
D bypass_two_reshape: remove node = ['/model.22/Concat_3_output_0_rs', '/model.22/Concat_3_output_0-rs']
D fuse_two_reshape: remove node = ['/model.22/Split_output_0-rs', '/model.22/dfl/Reshape_1']
D bypass_two_reshape: remove node = ['/model.22/Slice_output_0_rs', '/model.22/Slice_output_0-rs', '/model.22/Sub_output_0_rs#1', '/model.22/Sub_output_0_rs', '/model.22/Sub_output_0-rs', '/model.22/Slice_1_output_0_rs', '/model.22/Slice_1_output_0-rs', '/model.22/Add_1_output_0_rs#1', '/model.22/Add_1_output_0_rs', '/model.22/Add_1_output_0-rs', '/model.22/Add_2_output_0_rs', '/model.22/Add_2_output_0-rs', '/model.22/Div_1_output_0_rs', '/model.22/Div_1_output_0-rs']
D reduce_tp_in_mesh_concat: remove node = ['/model.22/Div_1_output_0_rs_rs', '/model.22/Sub_1_output_0_rs_rs', '/model.22/Concat_4_output_0-rs-rs']
D bypass_two_reshape: remove node = ['/model.22/Concat_4_output_0_rs', '/model.22/Concat_4_output_0-rs', '/model.22/Mul_2_output_0_rs', '/model.22/Mul_2_output_0-rs', '/model.22/Split_output_1_rs', '/model.22/Split_output_1-rs', '/model.22/Sigmoid_output_0_rs', '/model.22/Sigmoid_output_0-rs']
D reduce_tp_in_mesh_concat: remove node = ['/model.22/Mul_2_output_0_rs_rs', '/model.22/Sigmoid_output_0_rs_rs', 'output0-rs-rs']
D fuse_reshape_transpose: remove node = ['/model.22/dfl/Transpose'], add node = ['/model.22/dfl/Transpose']
D bypass_two_reshape: remove node = ['/model.22/Sub_1_output_0_rs', '/model.22/Sub_1_output_0-rs']
D convert_reshape_to_transpose: remove node = ['/model.22/dfl/Reshape_1_output_0_rs'], add node = ['/model.22/dfl/Reshape_1_output_0_rs']
D fold_constant ...
D fold_constant done.
D fuse_ops done.
D
W build: found outlier value, this may affect quantization accuracy
const name abs_mean abs_std outlier value
model.0.conv.weight 4.03 4.41 26.039
D sparse_weight ...
D sparse_weight done.
D
I GraphPreparing : 100%|████████████████████████████████████████| 176/176 [00:00<00:00, 1449.01it/s]
I Quantizating : 100%|████████████████████████████████████████████| 176/176 [00:03<00:00, 52.53it/s]
D
D quant_optimizer ...
D quant_optimizer results:
D adjust_tanh_sigmoid: ['/model.22/Sigmoid']
D adjust_concat_split: ['/model.22/Split', '/model.22/Concat_3', '/model.22/Concat', '/model.22/Concat_1', '/model.22/Concat_2']
D adjust_no_change_node: ['/model.22/Reshape_output_0_rs', '/model.22/Reshape_1_output_0_rs', '/model.22/Reshape_2_output_0_rs', '/model.9/m_2/MaxPool', '/model.9/m_1/MaxPool', '/model.9/m/MaxPool']
D quant_optimizer done.
D
D recover_const_share ...
D recover_const_share done.
D
W build: The default input dtype of 'images' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
W build: The default output dtype of 'output0' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
I rknn building ...
I RKNN: [13:38:22.064] compress = 0, conv_eltwise_activation_fuse = 1, global_fuse = 1, multi-core-model-mode = 7, output_optimize = 1, layout_match = 1, enable_argb_group = 0, pipeline_fuse = 1, enable_flash_attention = 0
I RKNN: librknnc version: 2.1.0 (967d001cc8@2024-08-07T11:32:45)
D RKNN: [13:38:22.114] RKNN is invoked
W RKNN: [13:38:22.299] Model initializer tensor data is empty, name: empty_placeholder_0
D RKNN: [13:38:22.301] >>>>>> start: rknn::RKNNExtractCustomOpAttrs
D RKNN: [13:38:22.301] <<<<<<<< end: rknn::RKNNExtractCustomOpAttrs
D RKNN: [13:38:22.301] >>>>>> start: rknn::RKNNSetOpTargetPass
D RKNN: [13:38:22.301] <<<<<<<< end: rknn::RKNNSetOpTargetPass
D RKNN: [13:38:22.301] >>>>>> start: rknn::RKNNBindNorm
D RKNN: [13:38:22.302] <<<<<<<< end: rknn::RKNNBindNorm
D RKNN: [13:38:22.302] >>>>>> start: rknn::RKNNEliminateQATDataConvert
D RKNN: [13:38:22.302] <<<<<<<< end: rknn::RKNNEliminateQATDataConvert
D RKNN: [13:38:22.302] >>>>>> start: rknn::RKNNTileGroupConv
D RKNN: [13:38:22.302] <<<<<<<< end: rknn::RKNNTileGroupConv
D RKNN: [13:38:22.302] >>>>>> start: rknn::RKNNAddConvBias
D RKNN: [13:38:22.302] <<<<<<<< end: rknn::RKNNAddConvBias
D RKNN: [13:38:22.302] >>>>>> start: rknn::RKNNTileChannel
D RKNN: [13:38:22.302] <<<<<<<< end: rknn::RKNNTileChannel
D RKNN: [13:38:22.302] >>>>>> start: rknn::RKNNPerChannelPrep
D RKNN: [13:38:22.303] <<<<<<<< end: rknn::RKNNPerChannelPrep
D RKNN: [13:38:22.303] >>>>>> start: rknn::RKNNBnQuant
D RKNN: [13:38:22.303] <<<<<<<< end: rknn::RKNNBnQuant
D RKNN: [13:38:22.303] >>>>>> start: rknn::RKNNFuseOptimizerPass
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNFuseOptimizerPass
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNTurnAutoPad
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNTurnAutoPad
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNInitRNNConst
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNInitRNNConst
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNInitCastConst
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNInitCastConst
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNMultiSurfacePass
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNMultiSurfacePass
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNReplaceConstantTensorPass
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNReplaceConstantTensorPass
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNSubgraphManager
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNSubgraphManager
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNSubgraphManager
D RKNN: [13:38:22.316] >>>>>> start: rknn::RKNNAddSecondaryNode
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNAddSecondaryNode
D RKNN: [13:38:22.316] <<<<<<<< end: rknn::RKNNSubgraphManager
D RKNN: [13:38:22.316] >>>>>> start: OpEmit
E RKNN: [13:38:22.316] REGTASK: The bit width of field value exceeds the limit, target: v2, offset: 0x500c, shift = 0, limit: 0x1fff, value: 0x20cf
W RKNN: [13:38:22.316] emitABC_T_BAC_regtask notch_addr overflow
E RKNN: [13:38:22.316] REGTASK: The bit width of field value exceeds the limit, target: v2, offset: 0x4038, shift = 0, limit: 0x1fff, value: 0x419f
E RKNN: [13:38:22.316] REGTASK: The bit width of field value exceeds the limit, target: v2, offset: 0x4038, shift = 16, limit: 0x1fff, value: 0x419f
W RKNN: [13:38:22.316] emitABC_T_BAC_regtask notch_addr overflow
W RKNN: [13:38:22.317] Transpose will fallback to CPU, because input shape has exceeded the max limit, height(4) * width(8400) = 33600, required product no larger than 16384!
E RKNN: [13:38:22.317] REGTASK: The bit width of field value exceeds the limit, target: v2, offset: 0x500c, shift = 0, limit: 0x1fff, value: 0x20cf
D RKNN: [13:38:22.317] <<<<<<<< end: OpEmit
D RKNN: [13:38:22.317] >>>>>> start: rknn::RKNNAddFirstConv
D RKNN: [13:38:22.317] <<<<<<<< end: rknn::RKNNAddFirstConv
D RKNN: [13:38:22.317] >>>>>> start: rknn::RKNNTilingPass
W RKNN: [13:38:22.318] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:38:22.318] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:38:22.318] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:38:22.318] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:38:22.318] Please help report this bug!
W RKNN: [13:38:22.318] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:38:22.318] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:38:22.318] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:38:22.318] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:38:22.318] Please help report this bug!
W RKNN: [13:38:22.318] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:38:22.318] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:38:22.318] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:38:22.318] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:38:22.318] Please help report this bug!
W RKNN: [13:38:22.318] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:38:22.318] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:38:22.318] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:38:22.318] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:38:22.318] Please help report this bug!
W RKNN: [13:38:22.318] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:38:22.318] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:38:22.318] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:38:22.318] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:38:22.318] Please help report this bug!
D RKNN: [13:38:22.318] <<<<<<<< end: rknn::RKNNTilingPass
D RKNN: [13:38:22.318] >>>>>> start: rknn::RKNNLayoutMatchPass
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:38:22.318] BroadcastLayoutConnFactory: op(Mul:/model.22/Mul_2) has undefined broadcast type, please check this op.
W RKNN: [13:38:22.318] BroadcastLayoutConnFactory: op(Mul:/model.22/Mul_2) has undefined broadcast type, please check this op.
W RKNN: [13:38:22.318] BroadcastLayoutConnFactory: op(Mul:/model.22/Mul_2) has undefined broadcast type, please check this op.
W RKNN: [13:38:22.318] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
D RKNN: [13:38:22.318] <<<<<<<< end: rknn::RKNNLayoutMatchPass
D RKNN: [13:38:22.318] >>>>>> start: rknn::RKNNAddSecondaryNode
D RKNN: [13:38:22.318] <<<<<<<< end: rknn::RKNNAddSecondaryNode
D RKNN: [13:38:22.318] >>>>>> start: rknn::RKNNAllocateConvCachePass
D RKNN: [13:38:22.318] <<<<<<<< end: rknn::RKNNAllocateConvCachePass
D RKNN: [13:38:22.318] >>>>>> start: OpEmit
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.340] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.341] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.341] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.342] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.342] finish initComputeZoneMapByStepsVector
D RKNN: [13:38:22.342] <<<<<<<< end: OpEmit
D RKNN: [13:38:22.342] >>>>>> start: rknn::RKNNSubGraphMemoryPlanPass
D RKNN: [13:38:22.342] >>>>>> start: rknn::RKNNSubGraphMemoryPlanPass
D RKNN: [13:38:22.342] <<<<<<<< end: rknn::RKNNSubGraphMemoryPlanPass
D RKNN: [13:38:22.342] >>>>>> start: rknn::RKNNOperatorIdGenPass
D RKNN: [13:38:22.342] <<<<<<<< end: rknn::RKNNOperatorIdGenPass
D RKNN: [13:38:22.343] <<<<<<<< end: rknn::RKNNSubGraphMemoryPlanPass
D RKNN: [13:38:22.343] >>>>>> start: rknn::RKNNProfileAnalysisPass
D RKNN: [13:38:22.343] <<<<<<<< end: rknn::RKNNProfileAnalysisPass
D RKNN: [13:38:22.343] >>>>>> start: rknn::RKNNOperatorIdGenPass
D RKNN: [13:38:22.343] <<<<<<<< end: rknn::RKNNOperatorIdGenPass
D RKNN: [13:38:22.343] >>>>>> start: rknn::RKNNWeightTransposePass
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_137 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_137 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_157 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_157 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_157 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_157 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_184 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_184 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_184 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_184 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_184 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_184 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_211 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_211 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_211 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_211 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.10/Constant_output_0 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor empty_placeholder_0 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.10/Constant_output_0 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor empty_placeholder_0 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_388 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor onnx::Split_388 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/dfl/Constant_output_0 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/dfl/Constant_output_0 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Slice_2sp_split need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Slice_2sp_split need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Reshape_output_0_rs_i1 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Reshape_output_0_rs_i1 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Reshape_1_output_0_rs_i1 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Reshape_1_output_0_rs_i1 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Reshape_2_output_0_rs_i1 need paramter qtype, type is set to float16 by default!
W RKNN: [13:38:22.451] Warning: Tensor /model.22/Reshape_2_output_0_rs_i1 need paramter qtype, type is set to float16 by default!
D RKNN: [13:38:22.451] <<<<<<<< end: rknn::RKNNWeightTransposePass
D RKNN: [13:38:22.451] >>>>>> start: rknn::RKNNCPUWeightTransposePass
D RKNN: [13:38:22.451] <<<<<<<< end: rknn::RKNNCPUWeightTransposePass
D RKNN: [13:38:22.451] >>>>>> start: rknn::RKNNModelBuildPass
D RKNN: [13:38:22.474] <<<<<<<< end: rknn::RKNNModelBuildPass
D RKNN: [13:38:22.474] >>>>>> start: rknn::RKNNModelRegCmdbuildPass
D RKNN: [13:38:22.478] --------------------------------------------------------------------------------------------------------------------------------------------------------------------------
D RKNN: [13:38:22.478] Network Layer Information Table
D RKNN: [13:38:22.478] --------------------------------------------------------------------------------------------------------------------------------------------------------------------------
D RKNN: [13:38:22.478] ID OpType DataType Target InputShape OutputShape Cycles(DDR/NPU/Total) RW(KB) FullName
D RKNN: [13:38:22.478] --------------------------------------------------------------------------------------------------------------------------------------------------------------------------
D RKNN: [13:38:22.478] 0 InputOperator INT8 CPU \ (1,3,640,640) 0/0/0 0 InputOperator:images
D RKNN: [13:38:22.478] 1 ConvExSwish INT8 NPU (1,3,640,640),(32,3,3,3),(32) (1,32,320,320) 190686/921600/921600 1204 Conv:/model.0/conv/Conv
D RKNN: [13:38:22.478] 2 ConvExSwish INT8 NPU (1,32,320,320),(64,32,3,3),(64) (1,64,160,160) 208597/460800/460800 3218 Conv:/model.1/conv/Conv
D RKNN: [13:38:22.478] 3 ConvExSwish INT8 NPU (1,64,160,160),(64,64,1,1),(64) (1,64,160,160) 138726/204800/204800 1604 Conv:/model.2/cv1/conv/Conv
D RKNN: [13:38:22.478] 4 Split INT8 NPU (1,64,160,160),(2) (1,32,160,160),... 0/0/0 1600 Split:/model.2/Split
D RKNN: [13:38:22.478] 5 ConvExSwish INT8 NPU (1,32,160,160),(32,32,3,3),(32) (1,32,160,160) 69666/230400/230400 809 Conv:/model.2/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 6 ConvExSwish INT8 NPU (1,32,160,160),(32,32,3,3),(32) (1,32,160,160) 69666/230400/230400 809 Conv:/model.2/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 7 Add INT8 NPU (1,32,160,160),(1,32,160,160) (1,32,160,160) 0/0/0 1600 Add:/model.2/m.0/Add
D RKNN: [13:38:22.478] 8 Concat INT8 NPU (1,32,160,160),(1,32,160,160),... (1,96,160,160) 0/0/0 2400 Concat:/model.2/Concat
D RKNN: [13:38:22.478] 9 ConvExSwish INT8 NPU (1,96,160,160),(64,96,1,1),(64) (1,64,160,160) 173445/204800/204800 2406 Conv:/model.2/cv2/conv/Conv
D RKNN: [13:38:22.478] 10 ConvExSwish INT8 NPU (1,64,160,160),(128,64,3,3),(128) (1,128,80,80) 107059/460800/460800 1673 Conv:/model.3/conv/Conv
D RKNN: [13:38:22.478] 11 ConvExSwish INT8 NPU (1,128,80,80),(128,128,1,1),(128) (1,128,80,80) 70002/102400/102400 817 Conv:/model.4/cv1/conv/Conv
D RKNN: [13:38:22.478] 12 Split INT8 NPU (1,128,80,80),(2) (1,64,80,80),... 0/0/0 800 Split:/model.4/Split
D RKNN: [13:38:22.478] 13 ConvExSwish INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 36213/230400/230400 436 Conv:/model.4/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 14 ConvExSwish INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 36213/230400/230400 436 Conv:/model.4/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 15 Add INT8 NPU (1,64,80,80),(1,64,80,80) (1,64,80,80) 0/0/0 800 Add:/model.4/m.0/Add
D RKNN: [13:38:22.478] 16 ConvExSwish INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 36213/230400/230400 436 Conv:/model.4/m.1/cv1/conv/Conv
D RKNN: [13:38:22.478] 17 ConvExSwish INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 36213/230400/230400 436 Conv:/model.4/m.1/cv2/conv/Conv
D RKNN: [13:38:22.478] 18 Add INT8 NPU (1,64,80,80),(1,64,80,80) (1,64,80,80) 0/0/0 800 Add:/model.4/m.1/Add
D RKNN: [13:38:22.478] 19 Concat INT8 NPU (1,64,80,80),(1,64,80,80),... (1,256,80,80) 0/0/0 1600 Concat:/model.4/Concat
D RKNN: [13:38:22.478] 20 ConvExSwish INT8 NPU (1,256,80,80),(128,256,1,1),(128) (1,128,80,80) 105327/204800/204800 1633 Conv:/model.4/cv2/conv/Conv
D RKNN: [13:38:22.478] 21 ConvExSwish INT8 NPU (1,128,80,80),(256,128,3,3),(256) (1,256,40,40) 64504/460800/460800 1090 Conv:/model.5/conv/Conv
D RKNN: [13:38:22.478] 22 ConvExSwish INT8 NPU (1,256,40,40),(256,256,1,1),(256) (1,256,40,40) 37490/102400/102400 466 Conv:/model.6/cv1/conv/Conv
D RKNN: [13:38:22.478] 23 Split INT8 NPU (1,256,40,40),(2) (1,128,40,40),... 0/0/0 400 Split:/model.6/Split
D RKNN: [13:38:22.478] 24 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.6/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 25 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.6/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 26 Add INT8 NPU (1,128,40,40),(1,128,40,40) (1,128,40,40) 0/0/0 400 Add:/model.6/m.0/Add
D RKNN: [13:38:22.478] 27 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.6/m.1/cv1/conv/Conv
D RKNN: [13:38:22.478] 28 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.6/m.1/cv2/conv/Conv
D RKNN: [13:38:22.478] 29 Add INT8 NPU (1,128,40,40),(1,128,40,40) (1,128,40,40) 0/0/0 400 Add:/model.6/m.1/Add
D RKNN: [13:38:22.478] 30 Concat INT8 NPU (1,128,40,40),(1,128,40,40),... (1,512,40,40) 0/0/0 800 Concat:/model.6/Concat
D RKNN: [13:38:22.478] 31 ConvExSwish INT8 NPU (1,512,40,40),(256,512,1,1),(256) (1,256,40,40) 57577/204800/204800 930 Conv:/model.6/cv2/conv/Conv
D RKNN: [13:38:22.478] 32 ConvExSwish INT8 NPU (1,256,40,40),(512,256,3,3),(512) (1,512,20,20) 76019/460800/460800 1556 Conv:/model.7/conv/Conv
D RKNN: [13:38:22.478] 33 ConvExSwish INT8 NPU (1,512,20,20),(512,512,1,1),(512) (1,512,20,20) 28572/102400/102400 460 Conv:/model.8/cv1/conv/Conv
D RKNN: [13:38:22.478] 34 Split INT8 NPU (1,512,20,20),(2) (1,256,20,20),... 0/0/0 200 Split:/model.8/Split
D RKNN: [13:38:22.478] 35 ConvExSwish INT8 NPU (1,256,20,20),(256,256,3,3),(256) (1,256,20,20) 33681/230400/230400 678 Conv:/model.8/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 36 ConvExSwish INT8 NPU (1,256,20,20),(256,256,3,3),(256) (1,256,20,20) 33681/230400/230400 678 Conv:/model.8/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 37 Add INT8 NPU (1,256,20,20),(1,256,20,20) (1,256,20,20) 0/0/0 200 Add:/model.8/m.0/Add
D RKNN: [13:38:22.478] 38 Concat INT8 NPU (1,256,20,20),(1,256,20,20),... (1,768,20,20) 0/0/0 300 Concat:/model.8/Concat
D RKNN: [13:38:22.478] 39 ConvExSwish INT8 NPU (1,768,20,20),(512,768,1,1),(512) (1,512,20,20) 38443/153600/153600 688 Conv:/model.8/cv2/conv/Conv
D RKNN: [13:38:22.478] 40 ConvExSwish INT8 NPU (1,512,20,20),(256,512,1,1),(256) (1,256,20,20) 18616/51200/51200 330 Conv:/model.9/cv1/conv/Conv
D RKNN: [13:38:22.478] 41 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 0/0/0 100 MaxPool:/model.9/m/MaxPool
D RKNN: [13:38:22.478] 42 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 0/0/0 100 MaxPool:/model.9/m_1/MaxPool
D RKNN: [13:38:22.478] 43 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 0/0/0 100 MaxPool:/model.9/m_2/MaxPool
D RKNN: [13:38:22.478] 44 Concat INT8 NPU (1,256,20,20),(1,256,20,20),... (1,1024,20,20) 0/0/0 400 Concat:/model.9/Concat
D RKNN: [13:38:22.478] 45 ConvExSwish INT8 NPU (1,1024,20,20),(512,1024,1,1),(512) (1,512,20,20) 48313/204800/204800 916 Conv:/model.9/cv2/conv/Conv
D RKNN: [13:38:22.478] 46 Resize INT8 NPU (1,512,20,20),(1),(4) (1,512,40,40) 0/0/0 200 Resize:/model.10/Resize
D RKNN: [13:38:22.478] 47 Concat INT8 NPU (1,512,40,40),(1,256,40,40) (1,768,40,40) 0/0/0 1200 Concat:/model.11/Concat
D RKNN: [13:38:22.478] 48 ConvExSwish INT8 NPU (1,768,40,40),(256,768,1,1),(256) (1,256,40,40) 77664/307200/307200 1394 Conv:/model.12/cv1/conv/Conv
D RKNN: [13:38:22.478] 49 Split INT8 NPU (1,256,40,40),(2) (1,128,40,40),... 0/0/0 400 Split:/model.12/Split
D RKNN: [13:38:22.478] 50 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.12/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 51 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.12/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 52 Concat INT8 NPU (1,128,40,40),(1,128,40,40),... (1,384,40,40) 0/0/0 600 Concat:/model.12/Concat
D RKNN: [13:38:22.478] 53 ConvExSwish INT8 NPU (1,384,40,40),(256,384,1,1),(256) (1,256,40,40) 47534/153600/153600 698 Conv:/model.12/cv2/conv/Conv
D RKNN: [13:38:22.478] 54 Resize INT8 NPU (1,256,40,40),(1),(4) (1,256,80,80) 0/0/0 400 Resize:/model.13/Resize
D RKNN: [13:38:22.478] 55 Concat INT8 NPU (1,256,80,80),(1,128,80,80) (1,384,80,80) 0/0/0 2400 Concat:/model.14/Concat
D RKNN: [13:38:22.478] 56 ConvExSwish INT8 NPU (1,384,80,80),(128,384,1,1),(128) (1,128,80,80) 140652/307200/307200 2449 Conv:/model.15/cv1/conv/Conv
D RKNN: [13:38:22.478] 57 Split INT8 NPU (1,128,80,80),(2) (1,64,80,80),... 0/0/0 800 Split:/model.15/Split
D RKNN: [13:38:22.478] 58 ConvExSwish INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 36213/230400/230400 436 Conv:/model.15/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 59 ConvExSwish INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 36213/230400/230400 436 Conv:/model.15/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 60 Concat INT8 NPU (1,64,80,80),(1,64,80,80),... (1,192,80,80) 0/0/0 1200 Concat:/model.15/Concat
D RKNN: [13:38:22.478] 61 ConvExSwish INT8 NPU (1,192,80,80),(128,192,1,1),(128) (1,128,80,80) 87664/153600/153600 1225 Conv:/model.15/cv2/conv/Conv
D RKNN: [13:38:22.478] 62 ConvExSwish INT8 NPU (1,128,80,80),(128,128,3,3),(128) (1,128,40,40) 49568/230400/230400 945 Conv:/model.16/conv/Conv
D RKNN: [13:38:22.478] 63 Concat INT8 NPU (1,128,40,40),(1,256,40,40) (1,384,40,40) 0/0/0 600 Concat:/model.17/Concat
D RKNN: [13:38:22.478] 64 ConvExSwish INT8 NPU (1,128,80,80),(64,128,3,3),(64) (1,64,80,80) 55088/460800/460800 872 Conv:/model.22/cv2.0/cv2.0.0/conv/Conv
D RKNN: [13:38:22.478] 65 ConvExSwish INT8 NPU (1,128,80,80),(128,128,3,3),(128) (1,128,80,80) 75543/921600/921600 945 Conv:/model.22/cv3.0/cv3.0.0/conv/Conv
D RKNN: [13:38:22.478] 66 ConvExSwish INT8 NPU (1,384,40,40),(256,384,1,1),(256) (1,256,40,40) 47534/153600/153600 698 Conv:/model.18/cv1/conv/Conv
D RKNN: [13:38:22.478] 67 Split INT8 NPU (1,256,40,40),(2) (1,128,40,40),... 0/0/0 400 Split:/model.18/Split
D RKNN: [13:38:22.478] 68 ConvExSwish INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 36213/230400/230400 436 Conv:/model.22/cv2.0/cv2.0.1/conv/Conv
D RKNN: [13:38:22.478] 69 Conv INT8 NPU (1,64,80,80),(64,64,1,1),(64) (1,64,80,80) 34828/51200/51200 404 Conv:/model.22/cv2.0/cv2.0.2/Conv
D RKNN: [13:38:22.478] 70 ConvExSwish INT8 NPU (1,128,80,80),(128,128,3,3),(128) (1,128,80,80) 75543/921600/921600 945 Conv:/model.22/cv3.0/cv3.0.1/conv/Conv
D RKNN: [13:38:22.478] 71 Conv INT8 NPU (1,128,80,80),(80,128,1,1),(80) (1,80,80,80) 56744/76800/76800 810 Conv:/model.22/cv3.0/cv3.0.2/Conv
D RKNN: [13:38:22.478] 72 Concat INT8 NPU (1,64,80,80),(1,80,80,80) (1,144,80,80) 0/0/0 900 Concat:/model.22/Concat
D RKNN: [13:38:22.478] 73 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.18/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 74 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.18/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 75 Concat INT8 NPU (1,128,40,40),(1,128,40,40),... (1,384,40,40) 0/0/0 600 Concat:/model.18/Concat
D RKNN: [13:38:22.478] 76 ConvExSwish INT8 NPU (1,384,40,40),(256,384,1,1),(256) (1,256,40,40) 47534/153600/153600 698 Conv:/model.18/cv2/conv/Conv
D RKNN: [13:38:22.478] 77 ConvExSwish INT8 NPU (1,256,40,40),(256,256,3,3),(256) (1,256,20,20) 46668/230400/230400 978 Conv:/model.19/conv/Conv
D RKNN: [13:38:22.478] 78 Concat INT8 NPU (1,256,20,20),(1,512,20,20) (1,768,20,20) 0/0/0 300 Concat:/model.20/Concat
D RKNN: [13:38:22.478] 79 ConvExSwish INT8 NPU (1,256,40,40),(64,256,3,3),(64) (1,64,40,40) 27901/230400/230400 544 Conv:/model.22/cv2.1/cv2.1.0/conv/Conv
D RKNN: [13:38:22.478] 80 ConvExSwish INT8 NPU (1,256,40,40),(128,256,3,3),(128) (1,128,40,40) 38486/460800/460800 689 Conv:/model.22/cv3.1/cv3.1.0/conv/Conv
D RKNN: [13:38:22.478] 81 ConvExSwish INT8 NPU (1,768,20,20),(512,768,1,1),(512) (1,512,20,20) 38443/153600/153600 688 Conv:/model.21/cv1/conv/Conv
D RKNN: [13:38:22.478] 82 Split INT8 NPU (1,512,20,20),(2) (1,256,20,20),... 0/0/0 200 Split:/model.21/Split
D RKNN: [13:38:22.478] 83 ConvExSwish INT8 NPU (1,64,40,40),(64,64,3,3),(64) (1,64,40,40) 10239/57600/57600 136 Conv:/model.22/cv2.1/cv2.1.1/conv/Conv
D RKNN: [13:38:22.478] 84 Conv INT8 NPU (1,64,40,40),(64,64,1,1),(64) (1,64,40,40) 8853/12800/12800 104 Conv:/model.22/cv2.1/cv2.1.2/Conv
D RKNN: [13:38:22.478] 85 ConvExSwish INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 23594/230400/230400 345 Conv:/model.22/cv3.1/cv3.1.1/conv/Conv
D RKNN: [13:38:22.478] 86 Conv INT8 NPU (1,128,40,40),(80,128,1,1),(80) (1,80,40,40) 14535/19200/19200 210 Conv:/model.22/cv3.1/cv3.1.2/Conv
D RKNN: [13:38:22.478] 87 Concat INT8 NPU (1,64,40,40),(1,80,40,40) (1,144,40,40) 0/0/0 225 Concat:/model.22/Concat_1
D RKNN: [13:38:22.478] 88 ConvExSwish INT8 NPU (1,256,20,20),(256,256,3,3),(256) (1,256,20,20) 33681/230400/230400 678 Conv:/model.21/m.0/cv1/conv/Conv
D RKNN: [13:38:22.478] 89 ConvExSwish INT8 NPU (1,256,20,20),(256,256,3,3),(256) (1,256,20,20) 33681/230400/230400 678 Conv:/model.21/m.0/cv2/conv/Conv
D RKNN: [13:38:22.478] 90 Concat INT8 NPU (1,256,20,20),(1,256,20,20),... (1,768,20,20) 0/0/0 300 Concat:/model.21/Concat
D RKNN: [13:38:22.478] 91 ConvExSwish INT8 NPU (1,768,20,20),(512,768,1,1),(512) (1,512,20,20) 38443/153600/153600 688 Conv:/model.21/cv2/conv/Conv
D RKNN: [13:38:22.478] 92 ConvExSwish INT8 NPU (1,512,20,20),(64,512,3,3),(64) (1,64,20,20) 22230/115200/115200 488 Conv:/model.22/cv2.2/cv2.2.0/conv/Conv
D RKNN: [13:38:22.478] 93 ConvExSwish INT8 NPU (1,512,20,20),(128,512,3,3),(128) (1,128,20,20) 35802/230400/230400 777 Conv:/model.22/cv3.2/cv3.2.0/conv/Conv
D RKNN: [13:38:22.478] 94 ConvExSwish INT8 NPU (1,64,20,20),(64,64,3,3),(64) (1,64,20,20) 3745/14400/14400 61 Conv:/model.22/cv2.2/cv2.2.1/conv/Conv
D RKNN: [13:38:22.478] 95 Conv INT8 NPU (1,64,20,20),(64,64,1,1),(64) (1,64,20,20) 2360/3200/3200 29 Conv:/model.22/cv2.2/cv2.2.2/Conv
D RKNN: [13:38:22.478] 96 ConvExSwish INT8 NPU (1,128,20,20),(128,128,3,3),(128) (1,128,20,20) 10607/57600/57600 195 Conv:/model.22/cv3.2/cv3.2.1/conv/Conv
D RKNN: [13:38:22.478] 97 Conv INT8 NPU (1,128,20,20),(80,128,1,1),(80) (1,80,20,20) 3983/4800/4800 60 Conv:/model.22/cv3.2/cv3.2.2/Conv
D RKNN: [13:38:22.478] 98 Concat INT8 NPU (1,64,20,20),(1,80,20,20) (1,144,20,20) 0/0/0 56 Concat:/model.22/Concat_2
D RKNN: [13:38:22.478] 99 Reshape INT8 NPU (1,144,80,80),(4) (1,144,1,6400) 0/0/0 900 Reshape:/model.22/Reshape_output_0_rs
D RKNN: [13:38:22.478] 100 Reshape INT8 NPU (1,144,40,40),(4) (1,144,1,1600) 0/0/0 225 Reshape:/model.22/Reshape_1_output_0_rs
D RKNN: [13:38:22.478] 101 Reshape INT8 NPU (1,144,20,20),(4) (1,144,1,400) 0/0/0 56 Reshape:/model.22/Reshape_2_output_0_rs
D RKNN: [13:38:22.478] 102 Concat INT8 NPU (1,144,1,6400),(1,144,1,1600),... (1,144,1,8400) 0/0/0 1181 Concat:/model.22/Concat_3
D RKNN: [13:38:22.478] 103 Split INT8 NPU (1,144,1,8400),(2) (1,64,1,8400),... 0/0/0 1181 Split:/model.22/Split
D RKNN: [13:38:22.478] 104 Sigmoid INT8 NPU (1,80,1,8400) (1,80,1,8400) 0/0/0 656 Sigmoid:/model.22/Sigmoid
D RKNN: [13:38:22.478] 105 Reshape INT8 NPU (1,64,1,8400),(4) (4,16,1,8400) 0/0/0 525 Reshape:/model.22/dfl/Reshape
D RKNN: [13:38:22.478] 106 Transpose INT8 NPU (4,16,1,8400) (1,16,4,8400) 0/0/0 525 Transpose:/model.22/dfl/Transpose
D RKNN: [13:38:22.478] 107 exSoftmax13 INT8 NPU (1,16,4,8400),(1,16,1,1) (1,16,4,8400) 0/0/0 525 exSoftmax13:/model.22/dfl/Softmax
D RKNN: [13:38:22.478] 108 Conv INT8 NPU (1,16,4,8400),(1,16,1,1),(1) (1,1,4,8400) 45463/134400/134400 525 Conv:/model.22/dfl/conv/Conv
D RKNN: [13:38:22.478] 109 Transpose INT8 NPU (1,1,4,8400) (1,4,1,8400) 0/0/0 525 Transpose:/model.22/dfl/Reshape_1_output_0_rs
D RKNN: [13:38:22.478] 110 Split INT8 NPU (1,4,1,8400),(2) (1,2,1,8400),... 0/0/0 131 Split:/model.22/Slice_2sp
D RKNN: [13:38:22.478] 111 Sub INT8 NPU (1,2,1,8400),(1,2,1,8400) (1,2,1,8400) 0/0/0 262 Sub:/model.22/Sub
D RKNN: [13:38:22.478] 112 Add INT8 NPU (1,2,1,8400),(1,2,1,8400) (1,2,1,8400) 0/0/0 262 Add:/model.22/Add_1
D RKNN: [13:38:22.478] 113 Sub INT8 NPU (1,2,1,8400),(1,2,1,8400) (1,2,1,8400) 0/0/0 262 Sub:/model.22/Sub_1
D RKNN: [13:38:22.478] 114 Add INT8 NPU (1,2,1,8400),(1,2,1,8400) (1,2,1,8400) 0/0/0 262 Add:/model.22/Add_2
D RKNN: [13:38:22.478] 115 Mul INT8 NPU (1,2,1,8400),(1,1,1,1) (1,2,1,8400) 0/0/0 131 Mul:/model.22/Div_1_2mul
D RKNN: [13:38:22.478] 116 Concat INT8 NPU (1,2,1,8400),(1,2,1,8400),(4,32,1,1) (1,4,1,8400) 0/0/0 262 Concat:/model.22/Concat_4
D RKNN: [13:38:22.478] 117 Mul INT8 NPU (1,4,1,8400),(1,1,1,8400) (1,4,1,8400) 0/0/0 262 Mul:/model.22/Mul_2
D RKNN: [13:38:22.478] 118 Concat INT8 NPU (1,4,1,8400),(1,80,1,8400),... (1,84,1,8400) 0/0/0 795 Concat:/model.22/Concat_5
D RKNN: [13:38:22.478] 119 Reshape INT8 CPU (1,84,1,8400),(3) (1,84,8400) 0/0/0 787 Reshape:output0-rs
D RKNN: [13:38:22.478] 120 OutputOperator INT8 CPU (1,84,8400) \ 0/0/0 689 OutputOperator:output0
D RKNN: [13:38:22.478] --------------------------------------------------------------------------------------------------------------------------------------------------------------------------
D RKNN: [13:38:22.479] <<<<<<<< end: rknn::RKNNModelRegCmdbuildPass
D RKNN: [13:38:22.479] >>>>>> start: rknn::RKNNModelExportPass
D RKNN: [13:38:22.479] Export RKNN model to /tmp/tmpq2sub6ck/check.rknn
D RKNN: [13:38:22.489] <<<<<<<< end: rknn::RKNNModelExportPass
D RKNN: [13:38:22.489] >>>>>> start: rknn::RKNNMemStatisticsPass
D RKNN: [13:38:22.489] ----------------------------------------------------------------------------------------------------------------------------------------------------------
D RKNN: [13:38:22.489] Feature Tensor Information Table
D RKNN: [13:38:22.489] ------------------------------------------------------------------------------------------------------------------------+---------------------------------
D RKNN: [13:38:22.489] ID User Tensor DataType DataFormat OrigShape NativeShape | [Start End) Size
D RKNN: [13:38:22.489] ------------------------------------------------------------------------------------------------------------------------+---------------------------------
D RKNN: [13:38:22.489] 1 ConvExSwish images INT8 NC1HWC2 (1,3,640,640) (1,1,640,640,3) | 0x00000000 0x0012c000 0x0012c000
D RKNN: [13:38:22.489] 2 ConvExSwish /model.0/act/Mul_output_0 INT8 NC1HWC2 (1,32,320,320) (1,2,320,320,16) | 0x0012c000 0x0044c000 0x00320000
D RKNN: [13:38:22.489] 3 ConvExSwish /model.1/act/Mul_output_0 INT8 NC1HWC2 (1,64,160,160) (1,4,160,160,16) | 0x0044c000 0x005dc000 0x00190000
D RKNN: [13:38:22.489] 4 Split /model.2/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,64,160,160) (1,4,160,160,16) | 0x00000000 0x00190000 0x00190000
D RKNN: [13:38:22.489] 5 ConvExSwish /model.2/Split_output_1 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x00190000 0x00258000 0x000c8000
D RKNN: [13:38:22.489] 6 ConvExSwish /model.2/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 7 Add /model.2/Split_output_1 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x00190000 0x00258000 0x000c8000
D RKNN: [13:38:22.489] 7 Add /model.2/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x000c8000 0x00190000 0x000c8000
D RKNN: [13:38:22.489] 8 Concat /model.2/Split_output_0 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x00258000 0x00320000 0x000c8000
D RKNN: [13:38:22.489] 8 Concat /model.2/Split_output_1 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x00190000 0x00258000 0x000c8000
D RKNN: [13:38:22.489] 8 Concat /model.2/m.0/Add_output_0 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 9 ConvExSwish /model.2/Concat_output_0 INT8 NC1HWC2 (1,96,160,160) (1,6,160,160,16) | 0x00320000 0x00578000 0x00258000
D RKNN: [13:38:22.489] 10 ConvExSwish /model.2/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,64,160,160) (1,4,160,160,16) | 0x00000000 0x00190000 0x00190000
D RKNN: [13:38:22.489] 11 ConvExSwish /model.3/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00190000 0x00258000 0x000c8000
D RKNN: [13:38:22.489] 12 Split /model.4/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 13 ConvExSwish /model.4/Split_output_1 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 14 ConvExSwish /model.4/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00000000 0x00064000 0x00064000
D RKNN: [13:38:22.489] 15 Add /model.4/Split_output_1 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 15 Add /model.4/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00064000 0x000c8000 0x00064000
D RKNN: [13:38:22.489] 16 ConvExSwish /model.4/m.0/Add_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00000000 0x00064000 0x00064000
D RKNN: [13:38:22.489] 17 ConvExSwish /model.4/m.1/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00064000 0x000c8000 0x00064000
D RKNN: [13:38:22.489] 18 Add /model.4/m.0/Add_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00000000 0x00064000 0x00064000
D RKNN: [13:38:22.489] 18 Add /model.4/m.1/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00190000 0x001f4000 0x00064000
D RKNN: [13:38:22.489] 19 Concat /model.4/Split_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x0012c000 0x00190000 0x00064000
D RKNN: [13:38:22.489] 19 Concat /model.4/Split_output_1 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 19 Concat /model.4/m.0/Add_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00000000 0x00064000 0x00064000
D RKNN: [13:38:22.489] 19 Concat /model.4/m.1/Add_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00064000 0x000c8000 0x00064000
D RKNN: [13:38:22.489] 20 ConvExSwish /model.4/Concat_output_0 INT8 NC1HWC2 (1,256,80,80) (1,16,80,80,16) | 0x00190000 0x00320000 0x00190000
D RKNN: [13:38:22.489] 21 ConvExSwish /model.4/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 22 ConvExSwish /model.5/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 23 Split /model.6/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x0012c000 0x00190000 0x00064000
D RKNN: [13:38:22.489] 24 ConvExSwish /model.6/Split_output_1 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x000c8000 0x000fa000 0x00032000
D RKNN: [13:38:22.489] 25 ConvExSwish /model.6/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 26 Add /model.6/Split_output_1 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x000c8000 0x000fa000 0x00032000
D RKNN: [13:38:22.489] 26 Add /model.6/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0015e000 0x00190000 0x00032000
D RKNN: [13:38:22.489] 27 ConvExSwish /model.6/m.0/Add_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 28 ConvExSwish /model.6/m.1/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0015e000 0x00190000 0x00032000
D RKNN: [13:38:22.489] 29 Add /model.6/m.0/Add_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 29 Add /model.6/m.1/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00190000 0x001c2000 0x00032000
D RKNN: [13:38:22.489] 30 Concat /model.6/Split_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x000fa000 0x0012c000 0x00032000
D RKNN: [13:38:22.489] 30 Concat /model.6/Split_output_1 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x000c8000 0x000fa000 0x00032000
D RKNN: [13:38:22.489] 30 Concat /model.6/m.0/Add_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 30 Concat /model.6/m.1/Add_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0015e000 0x00190000 0x00032000
D RKNN: [13:38:22.489] 31 ConvExSwish /model.6/Concat_output_0 INT8 NC1HWC2 (1,512,40,40) (1,32,40,40,16) | 0x00190000 0x00258000 0x000c8000
D RKNN: [13:38:22.489] 32 ConvExSwish /model.6/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 33 ConvExSwish /model.7/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 34 Split /model.8/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x0015e000 0x00190000 0x00032000
D RKNN: [13:38:22.489] 35 ConvExSwish /model.8/Split_output_1 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0012c000 0x00145000 0x00019000
D RKNN: [13:38:22.489] 36 ConvExSwish /model.8/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0015e000 0x00177000 0x00019000
D RKNN: [13:38:22.489] 37 Add /model.8/Split_output_1 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0012c000 0x00145000 0x00019000
D RKNN: [13:38:22.489] 37 Add /model.8/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00177000 0x00190000 0x00019000
D RKNN: [13:38:22.489] 38 Concat /model.8/Split_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00145000 0x0015e000 0x00019000
D RKNN: [13:38:22.489] 38 Concat /model.8/Split_output_1 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0012c000 0x00145000 0x00019000
D RKNN: [13:38:22.489] 38 Concat /model.8/m.0/Add_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0015e000 0x00177000 0x00019000
D RKNN: [13:38:22.489] 39 ConvExSwish /model.8/Concat_output_0 INT8 NC1HWC2 (1,768,20,20) (1,48,20,20,16) | 0x00177000 0x001c2000 0x0004b000
D RKNN: [13:38:22.489] 40 ConvExSwish /model.8/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 41 MaxPool /model.9/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0015e000 0x00177000 0x00019000
D RKNN: [13:38:22.489] 42 MaxPool /model.9/m/MaxPool_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0012c000 0x00145000 0x00019000
D RKNN: [13:38:22.489] 43 MaxPool /model.9/m_1/MaxPool_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00145000 0x0015e000 0x00019000
D RKNN: [13:38:22.489] 44 Concat /model.9/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0015e000 0x00177000 0x00019000
D RKNN: [13:38:22.489] 44 Concat /model.9/m/MaxPool_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0012c000 0x00145000 0x00019000
D RKNN: [13:38:22.489] 44 Concat /model.9/m_1/MaxPool_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00145000 0x0015e000 0x00019000
D RKNN: [13:38:22.489] 44 Concat /model.9/m_2/MaxPool_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00177000 0x00190000 0x00019000
D RKNN: [13:38:22.489] 45 ConvExSwish /model.9/Concat_output_0 INT8 NC1HWC2 (1,1024,20,20) (1,64,20,20,16) | 0x00190000 0x001f4000 0x00064000
D RKNN: [13:38:22.489] 46 Resize /model.9/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 47 Concat /model.10/Resize_output_0 INT8 NC1HWC2 (1,512,40,40) (1,32,40,40,16) | 0x0015e040 0x00226040 0x000c8000
D RKNN: [13:38:22.489] 47 Concat /model.6/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 48 ConvExSwish /model.11/Concat_output_0 INT8 NC1HWC2 (1,768,40,40) (1,48,40,40,16) | 0x00226040 0x00352040 0x0012c000
D RKNN: [13:38:22.489] 49 Split /model.12/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 50 ConvExSwish /model.12/Split_output_1 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00190000 0x001c2000 0x00032000
D RKNN: [13:38:22.489] 51 ConvExSwish /model.12/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x000c8000 0x000fa000 0x00032000
D RKNN: [13:38:22.489] 52 Concat /model.12/Split_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0015e000 0x00190000 0x00032000
D RKNN: [13:38:22.489] 52 Concat /model.12/Split_output_1 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00190000 0x001c2000 0x00032000
D RKNN: [13:38:22.489] 52 Concat /model.12/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x000fa000 0x0012c000 0x00032000
D RKNN: [13:38:22.489] 53 ConvExSwish /model.12/Concat_output_0 INT8 NC1HWC2 (1,384,40,40) (1,24,40,40,16) | 0x001c2000 0x00258000 0x00096000
D RKNN: [13:38:22.489] 54 Resize /model.12/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 55 Concat /model.13/Resize_output_0 INT8 NC1HWC2 (1,256,80,80) (1,16,80,80,16) | 0x0015e040 0x002ee040 0x00190000
D RKNN: [13:38:22.489] 55 Concat /model.4/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 56 ConvExSwish /model.14/Concat_output_0 INT8 NC1HWC2 (1,384,80,80) (1,24,80,80,16) | 0x002ee040 0x00546040 0x00258000
D RKNN: [13:38:22.489] 57 Split /model.15/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 58 ConvExSwish /model.15/Split_output_1 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x001c2000 0x00226000 0x00064000
D RKNN: [13:38:22.489] 59 ConvExSwish /model.15/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00000000 0x00064000 0x00064000
D RKNN: [13:38:22.489] 60 Concat /model.15/Split_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x0015e000 0x001c2000 0x00064000
D RKNN: [13:38:22.489] 60 Concat /model.15/Split_output_1 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x001c2000 0x00226000 0x00064000
D RKNN: [13:38:22.489] 60 Concat /model.15/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00064000 0x000c8000 0x00064000
D RKNN: [13:38:22.489] 61 ConvExSwish /model.15/Concat_output_0 INT8 NC1HWC2 (1,192,80,80) (1,12,80,80,16) | 0x00226000 0x00352000 0x0012c000
D RKNN: [13:38:22.489] 62 ConvExSwish /model.15/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 63 Concat /model.16/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0015e000 0x00190000 0x00032000
D RKNN: [13:38:22.489] 63 Concat /model.12/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 64 ConvExSwish /model.15/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 65 ConvExSwish /model.15/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00000000 0x000c8000 0x000c8000
D RKNN: [13:38:22.489] 66 ConvExSwish /model.17/Concat_output_0 INT8 NC1HWC2 (1,384,40,40) (1,24,40,40,16) | 0x00190000 0x00226000 0x00096000
D RKNN: [13:38:22.489] 67 Split /model.18/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x00000000 0x00064000 0x00064000
D RKNN: [13:38:22.489] 68 ConvExSwish /model.22/cv2.0/cv2.0.0/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 69 Conv /model.22/cv2.0/cv2.0.1/act/Mul_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x00000000 0x00064000 0x00064000
D RKNN: [13:38:22.489] 70 ConvExSwish /model.22/cv3.0/cv3.0.0/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x00226000 0x002ee000 0x000c8000
D RKNN: [13:38:22.489] 71 Conv /model.22/cv3.0/cv3.0.1/act/Mul_output_0 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x0015e000 0x00226000 0x000c8000
D RKNN: [13:38:22.489] 72 Concat /model.22/cv2.0/cv2.0.2/Conv_output_0 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x000c8000 0x0012c000 0x00064000
D RKNN: [13:38:22.489] 72 Concat /model.22/cv3.0/cv3.0.2/Conv_output_0 INT8 NC1HWC2 (1,80,80,80) (1,6,80,80,16) | 0x00226000 0x002bc000 0x00096000
D RKNN: [13:38:22.489] 73 ConvExSwish /model.18/Split_output_1 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00064000 0x00096000 0x00032000
D RKNN: [13:38:22.489] 74 ConvExSwish /model.18/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00000000 0x00032000 0x00032000
D RKNN: [13:38:22.489] 75 Concat /model.18/Split_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00096000 0x000c8000 0x00032000
D RKNN: [13:38:22.489] 75 Concat /model.18/Split_output_1 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00064000 0x00096000 0x00032000
D RKNN: [13:38:22.489] 75 Concat /model.18/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00032000 0x00064000 0x00032000
D RKNN: [13:38:22.489] 76 ConvExSwish /model.18/Concat_output_0 INT8 NC1HWC2 (1,384,40,40) (1,24,40,40,16) | 0x0015e000 0x001f4000 0x00096000
D RKNN: [13:38:22.489] 77 ConvExSwish /model.18/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x001f4000 0x00258000 0x00064000
D RKNN: [13:38:22.489] 78 Concat /model.19/act/Mul_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00258000 0x00271000 0x00019000
D RKNN: [13:38:22.489] 78 Concat /model.9/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x0012c000 0x0015e000 0x00032000
D RKNN: [13:38:22.489] 79 ConvExSwish /model.18/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x001f4000 0x00258000 0x00064000
D RKNN: [13:38:22.489] 80 ConvExSwish /model.18/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x001f4000 0x00258000 0x00064000
D RKNN: [13:38:22.489] 81 ConvExSwish /model.20/Concat_output_0 INT8 NC1HWC2 (1,768,20,20) (1,48,20,20,16) | 0x00271000 0x002bc000 0x0004b000
D RKNN: [13:38:22.489] 82 Split /model.21/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x00032000 0x00064000 0x00032000
D RKNN: [13:38:22.489] 83 ConvExSwish /model.22/cv2.1/cv2.1.0/act/Mul_output_0 INT8 NC1HWC2 (1,64,40,40) (1,4,40,40,16) | 0x00258000 0x00271000 0x00019000
D RKNN: [13:38:22.489] 84 Conv /model.22/cv2.1/cv2.1.1/act/Mul_output_0 INT8 NC1HWC2 (1,64,40,40) (1,4,40,40,16) | 0x002a3000 0x002bc000 0x00019000
D RKNN: [13:38:22.489] 85 ConvExSwish /model.22/cv3.1/cv3.1.0/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x00000000 0x00032000 0x00032000
D RKNN: [13:38:22.489] 86 Conv /model.22/cv3.1/cv3.1.1/act/Mul_output_0 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0004b000 0x0007d000 0x00032000
D RKNN: [13:38:22.489] 87 Concat /model.22/cv2.1/cv2.1.2/Conv_output_0 INT8 NC1HWC2 (1,64,40,40) (1,4,40,40,16) | 0x00032000 0x0004b000 0x00019000
D RKNN: [13:38:22.489] 87 Concat /model.22/cv3.1/cv3.1.2/Conv_output_0 INT8 NC1HWC2 (1,80,40,40) (1,6,40,40,16) | 0x00000000 0x00025800 0x00025800
D RKNN: [13:38:22.489] 88 ConvExSwish /model.21/Split_output_1 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0028a000 0x002a3000 0x00019000
D RKNN: [13:38:22.489] 89 ConvExSwish /model.21/m.0/cv1/act/Mul_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x002a3000 0x002bc000 0x00019000
D RKNN: [13:38:22.489] 90 Concat /model.21/Split_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00271000 0x0028a000 0x00019000
D RKNN: [13:38:22.489] 90 Concat /model.21/Split_output_1 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0028a000 0x002a3000 0x00019000
D RKNN: [13:38:22.489] 90 Concat /model.21/m.0/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x00000000 0x00019000 0x00019000
D RKNN: [13:38:22.489] 91 ConvExSwish /model.21/Concat_output_0 INT8 NC1HWC2 (1,768,20,20) (1,48,20,20,16) | 0x00083400 0x000ce400 0x0004b000
D RKNN: [13:38:22.489] 92 ConvExSwish /model.21/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x00000000 0x00032000 0x00032000
D RKNN: [13:38:22.489] 93 ConvExSwish /model.21/cv2/act/Mul_output_0 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x00000000 0x00032000 0x00032000
D RKNN: [13:38:22.489] 94 ConvExSwish /model.22/cv2.2/cv2.2.0/act/Mul_output_0 INT8 NC1HWC2 (1,64,20,20) (1,4,20,20,16) | 0x00032000 0x00038400 0x00006400
D RKNN: [13:38:22.489] 95 Conv /model.22/cv2.2/cv2.2.1/act/Mul_output_0 INT8 NC1HWC2 (1,64,20,20) (1,4,20,20,16) | 0x00044c00 0x0004b000 0x00006400
D RKNN: [13:38:22.489] 96 ConvExSwish /model.22/cv3.2/cv3.2.0/act/Mul_output_0 INT8 NC1HWC2 (1,128,20,20) (1,8,20,20,16) | 0x00038400 0x00044c00 0x0000c800
D RKNN: [13:38:22.489] 97 Conv /model.22/cv3.2/cv3.2.1/act/Mul_output_0 INT8 NC1HWC2 (1,128,20,20) (1,8,20,20,16) | 0x00006400 0x00012c00 0x0000c800
D RKNN: [13:38:22.489] 98 Concat /model.22/cv2.2/cv2.2.2/Conv_output_0 INT8 NC1HWC2 (1,64,20,20) (1,4,20,20,16) | 0x00000000 0x00006400 0x00006400
D RKNN: [13:38:22.489] 98 Concat /model.22/cv3.2/cv3.2.2/Conv_output_0 INT8 NC1HWC2 (1,80,20,20) (1,6,20,20,16) | 0x00012c00 0x0001c200 0x00009600
D RKNN: [13:38:22.489] 99 Reshape /model.22/Concat_output_0 INT8 NC1HWC2 (1,144,80,80) (1,9,80,80,16) | 0x002bc000 0x0039d000 0x000e1000
D RKNN: [13:38:22.489] 99 Reshape /model.22/Concat_output_0_exSecondary INT8 NC1HWC2 (1,144,80,80) (1,9,80,80,16) | 0x00083400 0x00164400 0x000e1000
D RKNN: [13:38:22.489] 100 Reshape /model.22/Concat_1_output_0 INT8 NC1HWC2 (1,144,40,40) (1,9,40,40,16) | 0x0004b000 0x00083400 0x00038400
D RKNN: [13:38:22.489] 100 Reshape /model.22/Concat_1_output_0_exSecondary INT8 NC1HWC2 (1,144,40,40) (1,9,40,40,16) | 0x00083400 0x000bb800 0x00038400
D RKNN: [13:38:22.489] 101 Reshape /model.22/Concat_2_output_0 INT8 NC1HWC2 (1,144,20,20) (1,9,20,20,16) | 0x0001c200 0x0002a300 0x0000e100
D RKNN: [13:38:22.489] 101 Reshape /model.22/Concat_2_output_0_exSecondary INT8 NC1HWC2 (1,144,20,20) (1,9,20,20,16) | 0x00000000 0x0000e100 0x0000e100
D RKNN: [13:38:22.489] 102 Concat /model.22/Reshape_output_0_rs INT8 NC1HWC2 (1,144,1,6400) (1,9,1,6400,16) | 0x002bc000 0x0039d000 0x000e1000
D RKNN: [13:38:22.489] 102 Concat /model.22/Reshape_1_output_0_rs INT8 NC1HWC2 (1,144,1,1600) (1,9,1,1600,16) | 0x0004b000 0x00083400 0x00038400
D RKNN: [13:38:22.489] 102 Concat /model.22/Reshape_2_output_0_rs INT8 NC1HWC2 (1,144,1,400) (1,9,1,400,16) | 0x0001c200 0x0002a300 0x0000e100
D RKNN: [13:38:22.489] 103 Split /model.22/Concat_3_output_0-rs INT8 NC1HWC2 (1,144,1,8400) (1,9,1,8400,16) | 0x00083400 0x001aa900 0x00127500
D RKNN: [13:38:22.489] 104 Sigmoid /model.22/Split_output_1-rs INT8 NC1HWC2 (1,80,1,8400) (1,5,1,8400,16) | 0x001aa900 0x0024ea00 0x000a4100
D RKNN: [13:38:22.489] 105 Reshape /model.22/Split_output_0-rs INT8 NC1HWC2 (1,64,1,8400) (1,4,1,8400,16) | 0x00000000 0x00083400 0x00083400
D RKNN: [13:38:22.489] 105 Reshape /model.22/Split_output_0-rs_exSecondary INT8 NC1HWC2 (1,64,1,8400) (1,4,1,8400,16) | 0x00127500 0x001aa900 0x00083400
D RKNN: [13:38:22.489] 106 Transpose /model.22/dfl/Reshape_output_0 INT8 NC1HWC2 (4,16,1,8400) (4,1,1,8400,16) | 0x00000000 0x00083400 0x00083400
D RKNN: [13:38:22.489] 107 exSoftmax13 /model.22/dfl/Transpose_output_0 INT8 NC1HWC2 (1,16,4,8400) (1,1,4,8400,16) | 0x00000000 0x00083400 0x00083400
D RKNN: [13:38:22.489] 107 exSoftmax13 /model.22/dfl/Transpose_output_0_exSecondary INT8 NC1HWC2 (1,16,4,8400) (1,7,4,8400,16) | 0x00127500 0x004be140 0x00396c40
D RKNN: [13:38:22.489] 108 Conv /model.22/dfl/Softmax_output_0 INT8 NC1HWC2 (1,16,4,8400) (1,1,4,8400,16) | 0x004be140 0x00541540 0x00083400
D RKNN: [13:38:22.489] 109 Transpose /model.22/dfl/conv/Conv_output_0 INT8 NC1HWC2 (1,1,4,8400) (1,2,4,8400,16) | 0x00127500 0x0022dd00 0x00106800
D RKNN: [13:38:22.489] 109 Transpose /model.22/dfl/conv/Conv_output_0_exSecondary INT8 NC1HWC2 (1,1,4,8400) (1,4,4,8400,16) | 0x0022dd00 0x0045ba00 0x0022dd00
D RKNN: [13:38:22.489] 110 Split /model.22/dfl/Reshape_1_output_0_rs INT8 NC1HWC2 (1,4,1,8400) (1,16,1,8400,16) | 0x0045ba00*0x00668a00 0x0020d000
D RKNN: [13:38:22.489] 111 Sub /model.22/Slice_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00041a00 0x00062700 0x00020d00
D RKNN: [13:38:22.489] 112 Add /model.22/Slice_1_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,2,1,8400,16) | 0x00000000 0x00041a00 0x00041a00
D RKNN: [13:38:22.489] 113 Sub /model.22/Add_1_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00041a00 0x00062700 0x00020d00
D RKNN: [13:38:22.489] 113 Sub /model.22/Sub_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00062700 0x00083400 0x00020d00
D RKNN: [13:38:22.489] 114 Add /model.22/Sub_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00062700 0x00083400 0x00020d00
D RKNN: [13:38:22.489] 114 Add /model.22/Add_1_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00041a00 0x00062700 0x00020d00
D RKNN: [13:38:22.489] 115 Mul /model.22/Add_2_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00020d00 0x00041a00 0x00020d00
D RKNN: [13:38:22.489] 116 Concat /model.22/Div_1_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00041a00 0x00062700 0x00020d00
D RKNN: [13:38:22.489] 116 Concat /model.22/Sub_1_output_0-rs INT8 NC1HWC2 (1,2,1,8400) (1,1,1,8400,16) | 0x00000000 0x00020d00 0x00020d00
D RKNN: [13:38:22.489] 116 Concat /model.22/Div_1_output_0-rs_exSecondary INT8 NC1HWC2 (1,4,1,8400) (1,2,1,8400,16) | 0x00127500 0x00168f00 0x00041a00
D RKNN: [13:38:22.489] 117 Mul /model.22/Concat_4_output_0-rs-rs INT8 NC1HWC2 (1,4,1,8400) (1,2,1,8400,16) | 0x00168f00 0x001aa900 0x00041a00
D RKNN: [13:38:22.489] 117 Mul /model.22/Concat_4_output_0-rs-rs_exSecondary INT8 NC1HWC2 (1,4,1,8400) (1,4,1,8400,16) | 0x00000000 0x00083400 0x00083400
D RKNN: [13:38:22.489] 118 Concat /model.22/Mul_2_output_0-rs INT8 NC1HWC2 (1,4,1,8400) (1,1,1,8400,16) | 0x00127500 0x00148200 0x00020d00
D RKNN: [13:38:22.489] 118 Concat /model.22/Sigmoid_output_0-rs INT8 NC1HWC2 (1,80,1,8400) (1,5,1,8400,16) | 0x00083400 0x00127500 0x000a4100
D RKNN: [13:38:22.489] 118 Concat /model.22/Mul_2_output_0-rs_exSecondary INT8 NC1HWC2 (1,84,1,8400) (1,6,1,8400,16) | 0x00148200 0x0020d000 0x000c4e00
D RKNN: [13:38:22.489] 119 Reshape output0-rs-rs INT8 NC1HWC2 (1,84,1,8400) (1,6,1,8400,16) | 0x0020d000 0x002d1e00 0x000c4e00
D RKNN: [13:38:22.489] 120 OutputOperator output0 INT8 UNDEFINED (1,84,8400) (1,84,8400) | 0x00000040 0x000ac480 0x000ac440
D RKNN: [13:38:22.489] ------------------------------------------------------------------------------------------------------------------------+---------------------------------
D RKNN: [13:38:22.489] --------------------------------------------------------------------------------------------------------------
D RKNN: [13:38:22.489] Const Tensor Information Table
D RKNN: [13:38:22.489] ----------------------------------------------------------------------------+---------------------------------
D RKNN: [13:38:22.489] ID User Tensor DataType OrigShape | [Start End) Size
D RKNN: [13:38:22.489] ----------------------------------------------------------------------------+---------------------------------
D RKNN: [13:38:22.489] 1 ConvExSwish model.0.conv.weight INT8 (32,3,3,3) | 0x00000000 0x00001200 0x00001200
D RKNN: [13:38:22.489] 1 ConvExSwish model.0.conv.bias INT32 (32) | 0x00001200 0x00001300 0x00000100
D RKNN: [13:38:22.489] 2 ConvExSwish model.1.conv.weight INT8 (64,32,3,3) | 0x00001300 0x00005b00 0x00004800
D RKNN: [13:38:22.489] 2 ConvExSwish model.1.conv.bias INT32 (64) | 0x00005b00 0x00005d00 0x00000200
D RKNN: [13:38:22.489] 3 ConvExSwish model.2.cv1.conv.weight INT8 (64,64,1,1) | 0x00005d00 0x00006d00 0x00001000
D RKNN: [13:38:22.489] 3 ConvExSwish model.2.cv1.conv.bias INT32 (64) | 0x00006d00 0x00006f00 0x00000200
D RKNN: [13:38:22.489] 4 Split onnx::Split_137 INT64 (2) | 0x00ab6a40 0x00ab6a50 0x00000010
D RKNN: [13:38:22.489] 5 ConvExSwish model.2.m.0.cv1.conv.weight INT8 (32,32,3,3) | 0x00008900 0x0000ad00 0x00002400
D RKNN: [13:38:22.489] 5 ConvExSwish model.2.m.0.cv1.conv.bias INT32 (32) | 0x0000ad00 0x0000ae00 0x00000100
D RKNN: [13:38:22.489] 6 ConvExSwish model.2.m.0.cv2.conv.weight INT8 (32,32,3,3) | 0x0000ae00 0x0000d200 0x00002400
D RKNN: [13:38:22.489] 6 ConvExSwish model.2.m.0.cv2.conv.bias INT32 (32) | 0x0000d200 0x0000d300 0x00000100
D RKNN: [13:38:22.489] 9 ConvExSwish model.2.cv2.conv.weight INT8 (64,96,1,1) | 0x00006f00 0x00008700 0x00001800
D RKNN: [13:38:22.489] 9 ConvExSwish model.2.cv2.conv.bias INT32 (64) | 0x00008700 0x00008900 0x00000200
D RKNN: [13:38:22.489] 10 ConvExSwish model.3.conv.weight INT8 (128,64,3,3) | 0x0000d300 0x0001f300 0x00012000
D RKNN: [13:38:22.489] 10 ConvExSwish model.3.conv.bias INT32 (128) | 0x0001f300 0x0001f700 0x00000400
D RKNN: [13:38:22.489] 11 ConvExSwish model.4.cv1.conv.weight INT8 (128,128,1,1) | 0x0001f700 0x00023700 0x00004000
D RKNN: [13:38:22.489] 11 ConvExSwish model.4.cv1.conv.bias INT32 (128) | 0x00023700 0x00023b00 0x00000400
D RKNN: [13:38:22.489] 12 Split onnx::Split_157 INT64 (2) | 0x00ab6a80 0x00ab6a90 0x00000010
D RKNN: [13:38:22.489] 13 ConvExSwish model.4.m.0.cv1.conv.weight INT8 (64,64,3,3) | 0x0002bf00 0x00034f00 0x00009000
D RKNN: [13:38:22.489] 13 ConvExSwish model.4.m.0.cv1.conv.bias INT32 (64) | 0x00034f00 0x00035100 0x00000200
D RKNN: [13:38:22.489] 14 ConvExSwish model.4.m.0.cv2.conv.weight INT8 (64,64,3,3) | 0x00035100 0x0003e100 0x00009000
D RKNN: [13:38:22.489] 14 ConvExSwish model.4.m.0.cv2.conv.bias INT32 (64) | 0x0003e100 0x0003e300 0x00000200
D RKNN: [13:38:22.489] 16 ConvExSwish model.4.m.1.cv1.conv.weight INT8 (64,64,3,3) | 0x0003e300 0x00047300 0x00009000
D RKNN: [13:38:22.489] 16 ConvExSwish model.4.m.1.cv1.conv.bias INT32 (64) | 0x00047300 0x00047500 0x00000200
D RKNN: [13:38:22.489] 17 ConvExSwish model.4.m.1.cv2.conv.weight INT8 (64,64,3,3) | 0x00047500 0x00050500 0x00009000
D RKNN: [13:38:22.489] 17 ConvExSwish model.4.m.1.cv2.conv.bias INT32 (64) | 0x00050500 0x00050700 0x00000200
D RKNN: [13:38:22.489] 20 ConvExSwish model.4.cv2.conv.weight INT8 (128,256,1,1) | 0x00023b00 0x0002bb00 0x00008000
D RKNN: [13:38:22.489] 20 ConvExSwish model.4.cv2.conv.bias INT32 (128) | 0x0002bb00 0x0002bf00 0x00000400
D RKNN: [13:38:22.489] 21 ConvExSwish model.5.conv.weight INT8 (256,128,3,3) | 0x00050700 0x00098700 0x00048000
D RKNN: [13:38:22.489] 21 ConvExSwish model.5.conv.bias INT32 (256) | 0x00098700 0x00098f00 0x00000800
D RKNN: [13:38:22.489] 22 ConvExSwish model.6.cv1.conv.weight INT8 (256,256,1,1) | 0x00098f00 0x000a8f00 0x00010000
D RKNN: [13:38:22.489] 22 ConvExSwish model.6.cv1.conv.bias INT32 (256) | 0x000a8f00 0x000a9700 0x00000800
D RKNN: [13:38:22.489] 23 Split onnx::Split_184 INT64 (2) | 0x00ab6ac0 0x00ab6ad0 0x00000010
D RKNN: [13:38:22.489] 24 ConvExSwish model.6.m.0.cv1.conv.weight INT8 (128,128,3,3) | 0x000c9f00 0x000edf00 0x00024000
D RKNN: [13:38:22.489] 24 ConvExSwish model.6.m.0.cv1.conv.bias INT32 (128) | 0x000edf00 0x000ee300 0x00000400
D RKNN: [13:38:22.489] 25 ConvExSwish model.6.m.0.cv2.conv.weight INT8 (128,128,3,3) | 0x000ee300 0x00112300 0x00024000
D RKNN: [13:38:22.489] 25 ConvExSwish model.6.m.0.cv2.conv.bias INT32 (128) | 0x00112300 0x00112700 0x00000400
D RKNN: [13:38:22.489] 27 ConvExSwish model.6.m.1.cv1.conv.weight INT8 (128,128,3,3) | 0x00112700 0x00136700 0x00024000
D RKNN: [13:38:22.489] 27 ConvExSwish model.6.m.1.cv1.conv.bias INT32 (128) | 0x00136700 0x00136b00 0x00000400
D RKNN: [13:38:22.489] 28 ConvExSwish model.6.m.1.cv2.conv.weight INT8 (128,128,3,3) | 0x00136b00 0x0015ab00 0x00024000
D RKNN: [13:38:22.489] 28 ConvExSwish model.6.m.1.cv2.conv.bias INT32 (128) | 0x0015ab00 0x0015af00 0x00000400
D RKNN: [13:38:22.489] 31 ConvExSwish model.6.cv2.conv.weight INT8 (256,512,1,1) | 0x000a9700 0x000c9700 0x00020000
D RKNN: [13:38:22.489] 31 ConvExSwish model.6.cv2.conv.bias INT32 (256) | 0x000c9700 0x000c9f00 0x00000800
D RKNN: [13:38:22.489] 32 ConvExSwish model.7.conv.weight INT8 (512,256,3,3) | 0x0015af00 0x0027af00 0x00120000
D RKNN: [13:38:22.489] 32 ConvExSwish model.7.conv.bias INT32 (512) | 0x0027af00 0x0027bf00 0x00001000
D RKNN: [13:38:22.489] 33 ConvExSwish model.8.cv1.conv.weight INT8 (512,512,1,1) | 0x0027bf00 0x002bbf00 0x00040000
D RKNN: [13:38:22.489] 33 ConvExSwish model.8.cv1.conv.bias INT32 (512) | 0x002bbf00 0x002bcf00 0x00001000
D RKNN: [13:38:22.489] 34 Split onnx::Split_211 INT64 (2) | 0x00ab6b00 0x00ab6b10 0x00000010
D RKNN: [13:38:22.489] 35 ConvExSwish model.8.m.0.cv1.conv.weight INT8 (256,256,3,3) | 0x0031df00 0x003adf00 0x00090000
D RKNN: [13:38:22.489] 35 ConvExSwish model.8.m.0.cv1.conv.bias INT32 (256) | 0x003adf00 0x003ae700 0x00000800
D RKNN: [13:38:22.489] 36 ConvExSwish model.8.m.0.cv2.conv.weight INT8 (256,256,3,3) | 0x003ae700 0x0043e700 0x00090000
D RKNN: [13:38:22.489] 36 ConvExSwish model.8.m.0.cv2.conv.bias INT32 (256) | 0x0043e700 0x0043ef00 0x00000800
D RKNN: [13:38:22.489] 39 ConvExSwish model.8.cv2.conv.weight INT8 (512,768,1,1) | 0x002bcf00 0x0031cf00 0x00060000
D RKNN: [13:38:22.489] 39 ConvExSwish model.8.cv2.conv.bias INT32 (512) | 0x0031cf00 0x0031df00 0x00001000
D RKNN: [13:38:22.489] 40 ConvExSwish model.9.cv1.conv.weight INT8 (256,512,1,1) | 0x0043ef00 0x0045ef00 0x00020000
D RKNN: [13:38:22.489] 40 ConvExSwish model.9.cv1.conv.bias INT32 (256) | 0x0045ef00 0x0045f700 0x00000800
D RKNN: [13:38:22.489] 45 ConvExSwish model.9.cv2.conv.weight INT8 (512,1024,1,1) | 0x0045f700 0x004df700 0x00080000
D RKNN: [13:38:22.489] 45 ConvExSwish model.9.cv2.conv.bias INT32 (512) | 0x004df700 0x004e0700 0x00001000
D RKNN: [13:38:22.489] 46 Resize empty_placeholder_0 FLOAT (1) | 0x00b19580 0x00b19680 0x00000100
D RKNN: [13:38:22.489] 46 Resize /model.10/Constant_output_0 FLOAT (4) | 0x00ab6b40 0x00ab6c40 0x00000100
D RKNN: [13:38:22.489] 48 ConvExSwish model.12.cv1.conv.weight INT8 (256,768,1,1) | 0x004e0700 0x00510700 0x00030000
D RKNN: [13:38:22.489] 48 ConvExSwish model.12.cv1.conv.bias INT32 (256) | 0x00510700 0x00510f00 0x00000800
D RKNN: [13:38:22.489] 49 Split onnx::Split_184 INT64 (2) | 0x00ab6ac0 0x00ab6ad0 0x00000010
D RKNN: [13:38:22.489] 50 ConvExSwish model.12.m.0.cv1.conv.weight INT8 (128,128,3,3) | 0x00529700 0x0054d700 0x00024000
D RKNN: [13:38:22.489] 50 ConvExSwish model.12.m.0.cv1.conv.bias INT32 (128) | 0x0054d700 0x0054db00 0x00000400
D RKNN: [13:38:22.489] 51 ConvExSwish model.12.m.0.cv2.conv.weight INT8 (128,128,3,3) | 0x0054db00 0x00571b00 0x00024000
D RKNN: [13:38:22.489] 51 ConvExSwish model.12.m.0.cv2.conv.bias INT32 (128) | 0x00571b00 0x00571f00 0x00000400
D RKNN: [13:38:22.489] 53 ConvExSwish model.12.cv2.conv.weight INT8 (256,384,1,1) | 0x00510f00 0x00528f00 0x00018000
D RKNN: [13:38:22.489] 53 ConvExSwish model.12.cv2.conv.bias INT32 (256) | 0x00528f00 0x00529700 0x00000800
D RKNN: [13:38:22.489] 54 Resize empty_placeholder_0 FLOAT (1) | 0x00b19580 0x00b19680 0x00000100
D RKNN: [13:38:22.489] 54 Resize /model.10/Constant_output_0 FLOAT (4) | 0x00ab6b40 0x00ab6c40 0x00000100
D RKNN: [13:38:22.489] 56 ConvExSwish model.15.cv1.conv.weight INT8 (128,384,1,1) | 0x00571f00 0x0057df00 0x0000c000
D RKNN: [13:38:22.489] 56 ConvExSwish model.15.cv1.conv.bias INT32 (128) | 0x0057df00 0x0057e300 0x00000400
D RKNN: [13:38:22.489] 57 Split onnx::Split_157 INT64 (2) | 0x00ab6a80 0x00ab6a90 0x00000010
D RKNN: [13:38:22.489] 58 ConvExSwish model.15.m.0.cv1.conv.weight INT8 (64,64,3,3) | 0x00584700 0x0058d700 0x00009000
D RKNN: [13:38:22.489] 58 ConvExSwish model.15.m.0.cv1.conv.bias INT32 (64) | 0x0058d700 0x0058d900 0x00000200
D RKNN: [13:38:22.489] 59 ConvExSwish model.15.m.0.cv2.conv.weight INT8 (64,64,3,3) | 0x0058d900 0x00596900 0x00009000
D RKNN: [13:38:22.489] 59 ConvExSwish model.15.m.0.cv2.conv.bias INT32 (64) | 0x00596900 0x00596b00 0x00000200
D RKNN: [13:38:22.489] 61 ConvExSwish model.15.cv2.conv.weight INT8 (128,192,1,1) | 0x0057e300 0x00584300 0x00006000
D RKNN: [13:38:22.489] 61 ConvExSwish model.15.cv2.conv.bias INT32 (128) | 0x00584300 0x00584700 0x00000400
D RKNN: [13:38:22.489] 62 ConvExSwish model.16.conv.weight INT8 (128,128,3,3) | 0x00596b00 0x005bab00 0x00024000
D RKNN: [13:38:22.489] 62 ConvExSwish model.16.conv.bias INT32 (128) | 0x005bab00 0x005baf00 0x00000400
D RKNN: [13:38:22.489] 64 ConvExSwish model.22.cv2.0.0.conv.weight INT8 (64,128,3,3) | 0x008a7f00 0x008b9f00 0x00012000
D RKNN: [13:38:22.489] 64 ConvExSwish model.22.cv2.0.0.conv.bias INT32 (64) | 0x008b9f00 0x008ba100 0x00000200
D RKNN: [13:38:22.489] 65 ConvExSwish model.22.cv3.0.0.conv.weight INT8 (128,128,3,3) | 0x00945100 0x00969100 0x00024000
D RKNN: [13:38:22.489] 65 ConvExSwish model.22.cv3.0.0.conv.bias INT32 (128) | 0x00969100 0x00969500 0x00000400
D RKNN: [13:38:22.489] 66 ConvExSwish model.18.cv1.conv.weight INT8 (256,384,1,1) | 0x005baf00 0x005d2f00 0x00018000
D RKNN: [13:38:22.489] 66 ConvExSwish model.18.cv1.conv.bias INT32 (256) | 0x005d2f00 0x005d3700 0x00000800
D RKNN: [13:38:22.489] 67 Split onnx::Split_184 INT64 (2) | 0x00ab6ac0 0x00ab6ad0 0x00000010
D RKNN: [13:38:22.489] 68 ConvExSwish model.22.cv2.0.1.conv.weight INT8 (64,64,3,3) | 0x008ba100 0x008c3100 0x00009000
D RKNN: [13:38:22.489] 68 ConvExSwish model.22.cv2.0.1.conv.bias INT32 (64) | 0x008c3100 0x008c3300 0x00000200
D RKNN: [13:38:22.489] 69 Conv model.22.cv2.0.2.weight INT8 (64,64,1,1) | 0x008c3300 0x008c4300 0x00001000
D RKNN: [13:38:22.489] 69 Conv model.22.cv2.0.2.bias INT32 (64) | 0x008c4300 0x008c4500 0x00000200
D RKNN: [13:38:22.489] 70 ConvExSwish model.22.cv3.0.1.conv.weight INT8 (128,128,3,3) | 0x00969500 0x0098d500 0x00024000
D RKNN: [13:38:22.489] 70 ConvExSwish model.22.cv3.0.1.conv.bias INT32 (128) | 0x0098d500 0x0098d900 0x00000400
D RKNN: [13:38:22.489] 71 Conv model.22.cv3.0.2.weight INT8 (80,128,1,1) | 0x0098d900 0x00990100 0x00002800
D RKNN: [13:38:22.489] 71 Conv model.22.cv3.0.2.bias INT32 (80) | 0x00990100 0x00990400 0x00000300
D RKNN: [13:38:22.489] 73 ConvExSwish model.18.m.0.cv1.conv.weight INT8 (128,128,3,3) | 0x005ebf00 0x0060ff00 0x00024000
D RKNN: [13:38:22.489] 73 ConvExSwish model.18.m.0.cv1.conv.bias INT32 (128) | 0x0060ff00 0x00610300 0x00000400
D RKNN: [13:38:22.489] 74 ConvExSwish model.18.m.0.cv2.conv.weight INT8 (128,128,3,3) | 0x00610300 0x00634300 0x00024000
D RKNN: [13:38:22.489] 74 ConvExSwish model.18.m.0.cv2.conv.bias INT32 (128) | 0x00634300 0x00634700 0x00000400
D RKNN: [13:38:22.489] 76 ConvExSwish model.18.cv2.conv.weight INT8 (256,384,1,1) | 0x005d3700 0x005eb700 0x00018000
D RKNN: [13:38:22.489] 76 ConvExSwish model.18.cv2.conv.bias INT32 (256) | 0x005eb700 0x005ebf00 0x00000800
D RKNN: [13:38:22.489] 77 ConvExSwish model.19.conv.weight INT8 (256,256,3,3) | 0x00634700 0x006c4700 0x00090000
D RKNN: [13:38:22.489] 77 ConvExSwish model.19.conv.bias INT32 (256) | 0x006c4700 0x006c4f00 0x00000800
D RKNN: [13:38:22.489] 79 ConvExSwish model.22.cv2.1.0.conv.weight INT8 (64,256,3,3) | 0x008c4500 0x008e8500 0x00024000
D RKNN: [13:38:22.489] 79 ConvExSwish model.22.cv2.1.0.conv.bias INT32 (64) | 0x008e8500 0x008e8700 0x00000200
D RKNN: [13:38:22.489] 80 ConvExSwish model.22.cv3.1.0.conv.weight INT8 (128,256,3,3) | 0x00990400 0x009d8400 0x00048000
D RKNN: [13:38:22.489] 80 ConvExSwish model.22.cv3.1.0.conv.bias INT32 (128) | 0x009d8400 0x009d8800 0x00000400
D RKNN: [13:38:22.489] 81 ConvExSwish model.21.cv1.conv.weight INT8 (512,768,1,1) | 0x006c4f00 0x00724f00 0x00060000
D RKNN: [13:38:22.489] 81 ConvExSwish model.21.cv1.conv.bias INT32 (512) | 0x00724f00 0x00725f00 0x00001000
D RKNN: [13:38:22.489] 82 Split onnx::Split_211 INT64 (2) | 0x00ab6b00 0x00ab6b10 0x00000010
D RKNN: [13:38:22.489] 83 ConvExSwish model.22.cv2.1.1.conv.weight INT8 (64,64,3,3) | 0x008e8700 0x008f1700 0x00009000
D RKNN: [13:38:22.489] 83 ConvExSwish model.22.cv2.1.1.conv.bias INT32 (64) | 0x008f1700 0x008f1900 0x00000200
D RKNN: [13:38:22.489] 84 Conv model.22.cv2.1.2.weight INT8 (64,64,1,1) | 0x008f1900 0x008f2900 0x00001000
D RKNN: [13:38:22.489] 84 Conv model.22.cv2.1.2.bias INT32 (64) | 0x008f2900 0x008f2b00 0x00000200
D RKNN: [13:38:22.489] 85 ConvExSwish model.22.cv3.1.1.conv.weight INT8 (128,128,3,3) | 0x009d8800 0x009fc800 0x00024000
D RKNN: [13:38:22.489] 85 ConvExSwish model.22.cv3.1.1.conv.bias INT32 (128) | 0x009fc800 0x009fcc00 0x00000400
D RKNN: [13:38:22.489] 86 Conv model.22.cv3.1.2.weight INT8 (80,128,1,1) | 0x009fcc00 0x009ff400 0x00002800
D RKNN: [13:38:22.489] 86 Conv model.22.cv3.1.2.bias INT32 (80) | 0x009ff400 0x009ff700 0x00000300
D RKNN: [13:38:22.489] 88 ConvExSwish model.21.m.0.cv1.conv.weight INT8 (256,256,3,3) | 0x00786f00 0x00816f00 0x00090000
D RKNN: [13:38:22.489] 88 ConvExSwish model.21.m.0.cv1.conv.bias INT32 (256) | 0x00816f00 0x00817700 0x00000800
D RKNN: [13:38:22.489] 89 ConvExSwish model.21.m.0.cv2.conv.weight INT8 (256,256,3,3) | 0x00817700 0x008a7700 0x00090000
D RKNN: [13:38:22.489] 89 ConvExSwish model.21.m.0.cv2.conv.bias INT32 (256) | 0x008a7700 0x008a7f00 0x00000800
D RKNN: [13:38:22.489] 91 ConvExSwish model.21.cv2.conv.weight INT8 (512,768,1,1) | 0x00725f00 0x00785f00 0x00060000
D RKNN: [13:38:22.489] 91 ConvExSwish model.21.cv2.conv.bias INT32 (512) | 0x00785f00 0x00786f00 0x00001000
D RKNN: [13:38:22.489] 92 ConvExSwish model.22.cv2.2.0.conv.weight INT8 (64,512,3,3) | 0x008f2b00 0x0093ab00 0x00048000
D RKNN: [13:38:22.489] 92 ConvExSwish model.22.cv2.2.0.conv.bias INT32 (64) | 0x0093ab00 0x0093ad00 0x00000200
D RKNN: [13:38:22.489] 93 ConvExSwish model.22.cv3.2.0.conv.weight INT8 (128,512,3,3) | 0x009ff700 0x00a8f700 0x00090000
D RKNN: [13:38:22.489] 93 ConvExSwish model.22.cv3.2.0.conv.bias INT32 (128) | 0x00a8f700 0x00a8fb00 0x00000400
D RKNN: [13:38:22.489] 94 ConvExSwish model.22.cv2.2.1.conv.weight INT8 (64,64,3,3) | 0x0093ad00 0x00943d00 0x00009000
D RKNN: [13:38:22.489] 94 ConvExSwish model.22.cv2.2.1.conv.bias INT32 (64) | 0x00943d00 0x00943f00 0x00000200
D RKNN: [13:38:22.489] 95 Conv model.22.cv2.2.2.weight INT8 (64,64,1,1) | 0x00943f00 0x00944f00 0x00001000
D RKNN: [13:38:22.489] 95 Conv model.22.cv2.2.2.bias INT32 (64) | 0x00944f00 0x00945100 0x00000200
D RKNN: [13:38:22.489] 96 ConvExSwish model.22.cv3.2.1.conv.weight INT8 (128,128,3,3) | 0x00a8fb00 0x00ab3b00 0x00024000
D RKNN: [13:38:22.489] 96 ConvExSwish model.22.cv3.2.1.conv.bias INT32 (128) | 0x00ab3b00 0x00ab3f00 0x00000400
D RKNN: [13:38:22.489] 97 Conv model.22.cv3.2.2.weight INT8 (80,128,1,1) | 0x00ab3f00 0x00ab6700 0x00002800
D RKNN: [13:38:22.489] 97 Conv model.22.cv3.2.2.bias INT32 (80) | 0x00ab6700 0x00ab6a00 0x00000300
D RKNN: [13:38:22.489] 99 Reshape /model.22/Reshape_output_0_rs_i1 INT64 (4) | 0x00b19400 0x00b19420 0x00000020
D RKNN: [13:38:22.489] 100 Reshape /model.22/Reshape_1_output_0_rs_i1 INT64 (4) | 0x00b19440 0x00b19460 0x00000020
D RKNN: [13:38:22.489] 101 Reshape /model.22/Reshape_2_output_0_rs_i1 INT64 (4) | 0x00b19480 0x00b194a0 0x00000020
D RKNN: [13:38:22.489] 103 Split onnx::Split_388 INT64 (2) | 0x00ab6c40 0x00ab6c50 0x00000010
D RKNN: [13:38:22.489] 105 Reshape /model.22/dfl/Constant_output_0 INT64 (4) | 0x00ab6c80 0x00ab6ca0 0x00000020
D RKNN: [13:38:22.489] 108 Conv model.22.dfl.conv.weight INT8 (1,16,1,1) | 0x00ab6a00 0x00ab6a20 0x00000020
D RKNN: [13:38:22.489] 108 Conv model.22.dfl.conv.weight_bias_0 INT32 (1) | 0x00b19680 0x00b19700 0x00000080
D RKNN: [13:38:22.489] 110 Split /model.22/Slice_2sp_split INT64 (2) | 0x00b193c0 0x00b193d0 0x00000010
D RKNN: [13:38:22.489] 111 Sub /model.22/Constant_9_output_0 INT8 (1,2,1,8400) | 0x00ab6cc0 0x00af86c0 0x00041a00
D RKNN: [13:38:22.489] 112 Add /model.22/Constant_9_output_0 INT8 (1,2,1,8400) | 0x00ab6cc0 0x00af86c0 0x00041a00
D RKNN: [13:38:22.489] 115 Mul /model.22/Div_1_2mul_i1 FLOAT (1,1,1,1) | 0x00b194c0 0x00b19540 0x00000080
D RKNN: [13:38:22.489] 117 Mul /model.22/Constant_12_output_0 INT8 (1,1,1,8400) | 0x00af86c0 0x00b193c0 0x00020d00
D RKNN: [13:38:22.489] 119 Reshape output0-rs_i1 INT64 (3) | 0x00b19540 0x00b19558 0x00000018
D RKNN: [13:38:22.489] ----------------------------------------------------------------------------+---------------------------------
D RKNN: [13:38:22.490] ----------------------------------------
D RKNN: [13:38:22.490] Total Internal Memory Size: 6562.5KB
D RKNN: [13:38:22.490] Total Weight Memory Size: 11367.8KB
D RKNN: [13:38:22.490] ----------------------------------------
D RKNN: [13:38:22.490] <<<<<<<< end: rknn::RKNNMemStatisticsPass
I rknn buiding done.
done
--> Export rknn model
done
I rknn-toolkit2 version: 2.0.0b0+9bab5682
--> Configuring model yolov8_small.onnx
done
--> Loading model
I It is recommended onnx opset 19, but your onnx model opset is 17!
I Model converted from pytorch, 'opset_version' should be set 19 in torch.onnx.export for successful convert!
I Loading : 100%|█████████████████████████████████████████████| 143/143 [00:00<00:00, 190650.18it/s]
done
--> Building model
D base_optimize ...
D base_optimize done.
D
D fold_constant ...
A module that was compiled using NumPy 1.x cannot be run in
NumPy 2.1.1 as it may crash. To support both 1.x and 2.x
versions of NumPy, modules must be compiled with NumPy 2.0.
Some module may need to rebuild instead e.g. with 'pybind11>=2.12'.
If you are a user of the module, the easiest solution will be to
downgrade to 'numpy<2' or try to upgrade the affected module.
We expect that some modules will need time to support NumPy 2.
Traceback (most recent call last): File "/home/vstepanov/RKNN_Compiller/compile.py", line 34, in <module>
ret = rknn.build(do_quantization=QUANTIZE_ON, dataset=DATASET)
File "/home/vstepanov/RKNN_Compiller/toolkit2.0-venv/lib/python3.11/site-packages/rknn/api/rknn.py", line 201, in build
return self.rknn_base.build(do_quantization=do_quantization, dataset=dataset, expand_batch_size=rknn_batch_size)
File "/home/vstepanov/RKNN_Compiller/toolkit2.0-venv/lib/python3.11/site-packages/onnxruntime/__init__.py", line 23, in <module>
from onnxruntime.capi._pybind_state import ExecutionMode # noqa: F401
File "/home/vstepanov/RKNN_Compiller/toolkit2.0-venv/lib/python3.11/site-packages/onnxruntime/capi/_pybind_state.py", line 32, in <module>
from .onnxruntime_pybind11_state import * # noqa
AttributeError: _ARRAY_API not found
ImportError: numpy.core.multiarray failed to import After downgrade numpy got another error I rknn-toolkit2 version: 2.0.0b0+9bab5682
--> Configuring model yolov8_small.onnx
done
--> Loading model
I It is recommended onnx opset 19, but your onnx model opset is 17!
I Model converted from pytorch, 'opset_version' should be set 19 in torch.onnx.export for successful convert!
I Loading : 100%|█████████████████████████████████████████████| 143/143 [00:00<00:00, 197492.75it/s]
done
--> Building model
D base_optimize ...
D base_optimize done.
D
D fold_constant ...
D fold_constant done.
D
D correct_ops ...
D correct_ops done.
D
D fuse_ops ...
D fuse_ops results:
D replace_exswish: remove node = ['/model.0/act/Sigmoid', '/model.0/act/Mul'], add node = ['/model.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.1/act/Sigmoid', '/model.1/act/Mul'], add node = ['/model.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/cv1/act/Sigmoid', '/model.2/cv1/act/Mul'], add node = ['/model.2/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/m.0/cv1/act/Sigmoid', '/model.2/m.0/cv1/act/Mul'], add node = ['/model.2/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/m.0/cv2/act/Sigmoid', '/model.2/m.0/cv2/act/Mul'], add node = ['/model.2/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.2/cv2/act/Sigmoid', '/model.2/cv2/act/Mul'], add node = ['/model.2/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.3/act/Sigmoid', '/model.3/act/Mul'], add node = ['/model.3/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/cv1/act/Sigmoid', '/model.4/cv1/act/Mul'], add node = ['/model.4/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.0/cv1/act/Sigmoid', '/model.4/m.0/cv1/act/Mul'], add node = ['/model.4/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.0/cv2/act/Sigmoid', '/model.4/m.0/cv2/act/Mul'], add node = ['/model.4/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.1/cv1/act/Sigmoid', '/model.4/m.1/cv1/act/Mul'], add node = ['/model.4/m.1/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/m.1/cv2/act/Sigmoid', '/model.4/m.1/cv2/act/Mul'], add node = ['/model.4/m.1/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.4/cv2/act/Sigmoid', '/model.4/cv2/act/Mul'], add node = ['/model.4/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.5/act/Sigmoid', '/model.5/act/Mul'], add node = ['/model.5/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/cv1/act/Sigmoid', '/model.6/cv1/act/Mul'], add node = ['/model.6/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.0/cv1/act/Sigmoid', '/model.6/m.0/cv1/act/Mul'], add node = ['/model.6/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.0/cv2/act/Sigmoid', '/model.6/m.0/cv2/act/Mul'], add node = ['/model.6/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.1/cv1/act/Sigmoid', '/model.6/m.1/cv1/act/Mul'], add node = ['/model.6/m.1/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/m.1/cv2/act/Sigmoid', '/model.6/m.1/cv2/act/Mul'], add node = ['/model.6/m.1/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.6/cv2/act/Sigmoid', '/model.6/cv2/act/Mul'], add node = ['/model.6/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.7/act/Sigmoid', '/model.7/act/Mul'], add node = ['/model.7/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/cv1/act/Sigmoid', '/model.8/cv1/act/Mul'], add node = ['/model.8/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/m.0/cv1/act/Sigmoid', '/model.8/m.0/cv1/act/Mul'], add node = ['/model.8/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/m.0/cv2/act/Sigmoid', '/model.8/m.0/cv2/act/Mul'], add node = ['/model.8/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.8/cv2/act/Sigmoid', '/model.8/cv2/act/Mul'], add node = ['/model.8/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.9/cv1/act/Sigmoid', '/model.9/cv1/act/Mul'], add node = ['/model.9/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.9/cv2/act/Sigmoid', '/model.9/cv2/act/Mul'], add node = ['/model.9/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/cv1/act/Sigmoid', '/model.12/cv1/act/Mul'], add node = ['/model.12/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/m.0/cv1/act/Sigmoid', '/model.12/m.0/cv1/act/Mul'], add node = ['/model.12/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/m.0/cv2/act/Sigmoid', '/model.12/m.0/cv2/act/Mul'], add node = ['/model.12/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.12/cv2/act/Sigmoid', '/model.12/cv2/act/Mul'], add node = ['/model.12/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/cv1/act/Sigmoid', '/model.15/cv1/act/Mul'], add node = ['/model.15/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/m.0/cv1/act/Sigmoid', '/model.15/m.0/cv1/act/Mul'], add node = ['/model.15/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/m.0/cv2/act/Sigmoid', '/model.15/m.0/cv2/act/Mul'], add node = ['/model.15/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.15/cv2/act/Sigmoid', '/model.15/cv2/act/Mul'], add node = ['/model.15/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.16/act/Sigmoid', '/model.16/act/Mul'], add node = ['/model.16/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/cv1/act/Sigmoid', '/model.18/cv1/act/Mul'], add node = ['/model.18/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/m.0/cv1/act/Sigmoid', '/model.18/m.0/cv1/act/Mul'], add node = ['/model.18/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/m.0/cv2/act/Sigmoid', '/model.18/m.0/cv2/act/Mul'], add node = ['/model.18/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.18/cv2/act/Sigmoid', '/model.18/cv2/act/Mul'], add node = ['/model.18/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.19/act/Sigmoid', '/model.19/act/Mul'], add node = ['/model.19/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/cv1/act/Sigmoid', '/model.21/cv1/act/Mul'], add node = ['/model.21/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/m.0/cv1/act/Sigmoid', '/model.21/m.0/cv1/act/Mul'], add node = ['/model.21/m.0/cv1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/m.0/cv2/act/Sigmoid', '/model.21/m.0/cv2/act/Mul'], add node = ['/model.21/m.0/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.21/cv2/act/Sigmoid', '/model.21/cv2/act/Mul'], add node = ['/model.21/cv2/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.2/cv2.2.0/act/Sigmoid', '/model.22/cv2.2/cv2.2.0/act/Mul'], add node = ['/model.22/cv2.2/cv2.2.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.2/cv2.2.1/act/Sigmoid', '/model.22/cv2.2/cv2.2.1/act/Mul'], add node = ['/model.22/cv2.2/cv2.2.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.2/cv3.2.0/act/Sigmoid', '/model.22/cv3.2/cv3.2.0/act/Mul'], add node = ['/model.22/cv3.2/cv3.2.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.2/cv3.2.1/act/Sigmoid', '/model.22/cv3.2/cv3.2.1/act/Mul'], add node = ['/model.22/cv3.2/cv3.2.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.1/cv2.1.0/act/Sigmoid', '/model.22/cv2.1/cv2.1.0/act/Mul'], add node = ['/model.22/cv2.1/cv2.1.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.1/cv2.1.1/act/Sigmoid', '/model.22/cv2.1/cv2.1.1/act/Mul'], add node = ['/model.22/cv2.1/cv2.1.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.1/cv3.1.0/act/Sigmoid', '/model.22/cv3.1/cv3.1.0/act/Mul'], add node = ['/model.22/cv3.1/cv3.1.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.1/cv3.1.1/act/Sigmoid', '/model.22/cv3.1/cv3.1.1/act/Mul'], add node = ['/model.22/cv3.1/cv3.1.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.0/cv2.0.0/act/Sigmoid', '/model.22/cv2.0/cv2.0.0/act/Mul'], add node = ['/model.22/cv2.0/cv2.0.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv2.0/cv2.0.1/act/Sigmoid', '/model.22/cv2.0/cv2.0.1/act/Mul'], add node = ['/model.22/cv2.0/cv2.0.1/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.0/cv3.0.0/act/Sigmoid', '/model.22/cv3.0/cv3.0.0/act/Mul'], add node = ['/model.22/cv3.0/cv3.0.0/act/Sigmoid_2swish']
D replace_exswish: remove node = ['/model.22/cv3.0/cv3.0.1/act/Sigmoid', '/model.22/cv3.0/cv3.0.1/act/Mul'], add node = ['/model.22/cv3.0/cv3.0.1/act/Sigmoid_2swish']
D replace_parallel_slice_by_split: remove node = ['/model.22/Slice', '/model.22/Slice_1'], add node = ['/model.22/Slice_2split']
D unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Concat_3_0_unsqueeze0', '/model.22/Concat_3_1_unsqueeze0', '/model.22/Concat_3_2_unsqueeze0', '/model.22/Concat_3_0_unsqueeze1']
D unsqueeze_to_4d_split: remove node = [], add node = ['/model.22/Split_0_unsqueeze0', '/model.22/Split_0_unsqueeze1', '/model.22/Split_1_unsqueeze1']
D convert_softmax_to_exsoftmax13: remove node = ['/model.22/dfl/Softmax'], add node = ['/model.22/dfl/Softmax']
D unsqueeze_to_4d_split: remove node = [], add node = ['/model.22/Slice_2split_0_unsqueeze0', '/model.22/Slice_2split_0_unsqueeze1', '/model.22/Slice_2split_1_unsqueeze1']
D unsqueeze_to_4d_sub: remove node = [], add node = ['/model.22/Sub_1_unsqueeze0', '/model.22/Sub_0_unsqueeze1']
D unsqueeze_to_4d_add: remove node = [], add node = ['/model.22/Add_1_1_unsqueeze0', '/model.22/Add_1_0_unsqueeze1']
D unsqueeze_to_4d_add: remove node = [], add node = ['/model.22/Add_2_0_unsqueeze0', '/model.22/Add_2_1_unsqueeze0', '/model.22/Add_2_0_unsqueeze1']
D convert_div_to_mul: remove node = ['/model.22/Div_1'], add node = ['/model.22/Div_1_2mul']
D unsqueeze_to_4d_sub: remove node = [], add node = ['/model.22/Sub_1_0_unsqueeze0', '/model.22/Sub_1_1_unsqueeze0', '/model.22/Sub_1_0_unsqueeze1']
D unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Concat_4_0_unsqueeze0', '/model.22/Concat_4_1_unsqueeze0', '/model.22/Concat_4_0_unsqueeze1']
D unsqueeze_to_4d_mul: remove node = [], add node = ['/model.22/Mul_2_0_unsqueeze0', '/model.22/Mul_2_0_unsqueeze1']
D unsqueeze_to_4d_sigmoid: remove node = [], add node = ['/model.22/Sigmoid_0_unsqueeze0', '/model.22/Sigmoid_0_unsqueeze1']
D unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Concat_5_0_unsqueeze0', '/model.22/Concat_5_1_unsqueeze0', '/model.22/Concat_5_0_unsqueeze1']
D fuse_two_reshape: remove node = ['/model.22/Reshape_2', '/model.22/Reshape_1', '/model.22/Reshape', '/model.22/Concat_3_0_unsqueeze1', '/model.22/Split_0_unsqueeze1', '/model.22/dfl/Reshape_1', '/model.22/Slice_2split_0_unsqueeze1']
D remove_parallel_reshape: remove node = ['/model.22/Add_2_0_unsqueeze0']
D fuse_two_reshape: remove node = ['/model.22/Slice_2split_1_unsqueeze1']
D remove_parallel_reshape: remove node = ['/model.22/Add_2_1_unsqueeze0']
D unsqueeze_to_4d_mul: remove node = [], add node = ['/model.22/Div_1_2mul_0_unsqueeze0', '/model.22/Div_1_2mul_0_unsqueeze1']
D swap_concat_axis_avoid_channel_concat: remove node = [], add node = ['/model.22/Concat_4_swap_concat_reshape_i0_out', '/model.22/Concat_4_swap_concat_reshape_i1_out', '/model.22/Concat_4_swap_concat_reshape_o0_out']
D fuse_two_reshape: remove node = ['/model.22/Concat_4_0_unsqueeze1']
D input_align_4D_mul: remove node = ['/model.22/Mul_2'], add node = ['/model.22/Mul_2']
D fuse_two_reshape: remove node = ['/model.22/Mul_2_0_unsqueeze1', '/model.22/Split_1_unsqueeze1', '/model.22/Sigmoid_0_unsqueeze1']
D swap_concat_axis_avoid_channel_concat: remove node = [], add node = ['/model.22/Concat_5_swap_concat_reshape_i0_out', '/model.22/Concat_5_swap_concat_reshape_i1_out', '/model.22/Concat_5_swap_concat_reshape_o0_out']
D remove_invalid_reshape: remove node = ['/model.22/Split_0_unsqueeze0', '/model.22/Sub_1_unsqueeze0']
D fuse_two_reshape: remove node = ['/model.22/Sub_0_unsqueeze1']
D remove_invalid_reshape: remove node = ['/model.22/Add_1_1_unsqueeze0']
D fuse_two_reshape: remove node = ['/model.22/Add_1_0_unsqueeze1', '/model.22/Sub_1_0_unsqueeze1', '/model.22/Add_2_0_unsqueeze1', '/model.22/Div_1_2mul_0_unsqueeze1', '/model.22/Concat_4_swap_concat_reshape_o0_out', '/model.22/Concat_5_0_unsqueeze0']
D remove_invalid_reshape: remove node = ['/model.22/Sigmoid_0_unsqueeze0']
D fuse_two_reshape: remove node = ['/model.22/Concat_5_1_unsqueeze0', '/model.22/Concat_5_swap_concat_reshape_o0_out']
D remove_invalid_reshape: remove node = ['/model.22/Sub_1_1_unsqueeze0', '/model.22/Sub_1_0_unsqueeze0']
D fuse_two_reshape: remove node = ['/model.22/Concat_4_1_unsqueeze0']
D remove_invalid_reshape: remove node = ['/model.22/Div_1_2mul_0_unsqueeze0']
D fuse_two_reshape: remove node = ['/model.22/Concat_4_0_unsqueeze0']
D replace_batch_shuffle_transpose_by_gather_after_reshape: remove node = ['/model.22/dfl/Reshape', '/model.22/dfl/Transpose'], add node = ['/model.22/dfl/Reshape', '/model.22/dfl/Transpose_2_gather']
D convert_reshape_to_transpose: remove node = ['/model.22/Slice_2split_0_unsqueeze0'], add node = ['/model.22/Slice_2split_0_unsqueeze0']
D convert_reshape_to_transpose: remove node = ['/model.22/Concat_4_swap_concat_reshape_i1_out'], add node = ['/model.22/Concat_4_swap_concat_reshape_i1_out']
D convert_reshape_to_transpose: remove node = ['/model.22/Concat_4_swap_concat_reshape_i0_out'], add node = ['/model.22/Concat_4_swap_concat_reshape_i0_out']
D reduce_transpose_op_around_concat: remove node = ['/model.22/Concat_4'], add node = ['/model.22/Concat_4_swap_concat_reshape_i0_out_tp_4/model.22/Concat_4', '/model.22/Concat_4_swap_concat_reshape_i1_out_tp_4/model.22/Concat_4', '/model.22/Concat_4', '/model.22/Concat_4_output_0_shape4_tp_4/model.22/Concat_4']
D convert_reshape_to_transpose: remove node = ['/model.22/Mul_2_0_unsqueeze0'], add node = ['/model.22/Mul_2_0_unsqueeze0']
D convert_reshape_to_transpose: remove node = ['/model.22/Concat_5_swap_concat_reshape_i0_out'], add node = ['/model.22/Concat_5_swap_concat_reshape_i0_out']
D convert_reshape_to_transpose: remove node = ['/model.22/Concat_5_swap_concat_reshape_i1_out'], add node = ['/model.22/Concat_5_swap_concat_reshape_i1_out']
D reduce_transpose_op_around_concat: remove node = ['/model.22/Concat_5'], add node = ['/model.22/Concat_5_swap_concat_reshape_i0_out_tp_4/model.22/Concat_5', '/model.22/Concat_5_swap_concat_reshape_i1_out_tp_4/model.22/Concat_5', '/model.22/Concat_5', 'output0_shape4_tp_4/model.22/Concat_5']
D convert_transpose_to_reshape: remove node = ['/model.22/Slice_2split_0_unsqueeze0'], add node = ['/model.22/Slice_2split_0_unsqueeze0_2reshape']
D bypass_two_transpose: remove node = ['/model.22/Concat_4_swap_concat_reshape_i1_out_tp_4/model.22/Concat_4', '/model.22/Concat_4_swap_concat_reshape_i1_out', '/model.22/Concat_4_swap_concat_reshape_i0_out_tp_4/model.22/Concat_4', '/model.22/Concat_4_swap_concat_reshape_i0_out', '/model.22/Mul_2_0_unsqueeze0', '/model.22/Concat_4_output_0_shape4_tp_4/model.22/Concat_4', '/model.22/Concat_5_swap_concat_reshape_i0_out_tp_4/model.22/Concat_5', '/model.22/Concat_5_swap_concat_reshape_i0_out', '/model.22/Concat_5_swap_concat_reshape_i1_out_tp_4/model.22/Concat_5', '/model.22/Concat_5_swap_concat_reshape_i1_out']
D convert_transpose_to_reshape: remove node = ['output0_shape4_tp_4/model.22/Concat_5'], add node = ['output0_shape4_tp_4/model.22/Concat_5_2reshape']
D convert_reshape_to_transpose: remove node = ['/model.22/Slice_2split_0_unsqueeze0_2reshape'], add node = ['/model.22/Slice_2split_0_unsqueeze0_2reshape']
D fuse_two_reshape: remove node = ['output0_shape4_tp_4/model.22/Concat_5_2reshape']
D fold_constant ...
D fold_constant done.
D fuse_ops done.
D
W build: found outlier value, this may affect quantization accuracy
const name abs_mean abs_std outlier value
model.0.conv.weight 4.03 4.41 26.039
D sparse_weight ...
D sparse_weight done.
D
I GraphPreparing : 100%|████████████████████████████████████████| 176/176 [00:00<00:00, 1900.26it/s]
I Quantizating : 100%|███████████████████████████████████████████| 176/176 [00:00<00:00, 480.21it/s]
D
D quant_optimizer ...
D quant_optimizer results:
D adjust_tanh_sigmoid: ['/model.22/Sigmoid']
D adjust_concat_split: ['/model.22/Split', '/model.22/Concat_3', '/model.22/Concat', '/model.22/Concat_1', '/model.22/Concat_2']
D adjust_no_change_node: ['/model.22/Concat_3_0_unsqueeze0', '/model.22/Concat_3_1_unsqueeze0', '/model.22/Concat_3_2_unsqueeze0', '/model.9/m_2/MaxPool', '/model.9/m_1/MaxPool', '/model.9/m/MaxPool']
D quant_optimizer done.
D
D recover_const_share ...
D recover_const_share done.
D
W build: The default input dtype of 'images' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
W build: The default output dtype of 'output0' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
I rknn building ...
I RKNN: [13:48:42.362] compress = 0, conv_eltwise_activation_fuse = 1, global_fuse = 1, multi-core-model-mode = 7, output_optimize = 1, layout_match = 1, enable_argb_group = 0
I RKNN: librknnc version: 2.0.0b0 (35a6907d79@2024-03-24T02:34:11)
D RKNN: [13:48:42.412] RKNN is invoked
W RKNN: [13:48:42.597] Model initializer tensor data is empty, name: empty_placeholder_0
D RKNN: [13:48:42.599] >>>>>> start: rknn::RKNNExtractCustomOpAttrs
D RKNN: [13:48:42.600] <<<<<<<< end: rknn::RKNNExtractCustomOpAttrs
D RKNN: [13:48:42.600] >>>>>> start: rknn::RKNNSetOpTargetPass
D RKNN: [13:48:42.600] <<<<<<<< end: rknn::RKNNSetOpTargetPass
D RKNN: [13:48:42.600] >>>>>> start: rknn::RKNNBindNorm
D RKNN: [13:48:42.600] <<<<<<<< end: rknn::RKNNBindNorm
D RKNN: [13:48:42.600] >>>>>> start: rknn::RKNNAddFirstConv
D RKNN: [13:48:42.600] <<<<<<<< end: rknn::RKNNAddFirstConv
D RKNN: [13:48:42.600] >>>>>> start: rknn::RKNNEliminateQATDataConvert
D RKNN: [13:48:42.600] <<<<<<<< end: rknn::RKNNEliminateQATDataConvert
D RKNN: [13:48:42.600] >>>>>> start: rknn::RKNNTileGroupConv
D RKNN: [13:48:42.600] <<<<<<<< end: rknn::RKNNTileGroupConv
D RKNN: [13:48:42.600] >>>>>> start: rknn::RKNNAddConvBias
D RKNN: [13:48:42.601] <<<<<<<< end: rknn::RKNNAddConvBias
D RKNN: [13:48:42.601] >>>>>> start: rknn::RKNNTileChannel
D RKNN: [13:48:42.601] <<<<<<<< end: rknn::RKNNTileChannel
D RKNN: [13:48:42.601] >>>>>> start: rknn::RKNNPerChannelPrep
D RKNN: [13:48:42.601] <<<<<<<< end: rknn::RKNNPerChannelPrep
D RKNN: [13:48:42.601] >>>>>> start: rknn::RKNNBnQuant
D RKNN: [13:48:42.601] <<<<<<<< end: rknn::RKNNBnQuant
D RKNN: [13:48:42.601] >>>>>> start: rknn::RKNNFuseOptimizerPass
D RKNN: [13:48:42.613] <<<<<<<< end: rknn::RKNNFuseOptimizerPass
D RKNN: [13:48:42.613] >>>>>> start: rknn::RKNNTurnAutoPad
D RKNN: [13:48:42.613] <<<<<<<< end: rknn::RKNNTurnAutoPad
D RKNN: [13:48:42.613] >>>>>> start: rknn::RKNNInitRNNConst
D RKNN: [13:48:42.613] <<<<<<<< end: rknn::RKNNInitRNNConst
D RKNN: [13:48:42.613] >>>>>> start: rknn::RKNNInitCastConst
D RKNN: [13:48:42.613] <<<<<<<< end: rknn::RKNNInitCastConst
D RKNN: [13:48:42.613] >>>>>> start: rknn::RKNNMultiSurfacePass
D RKNN: [13:48:42.613] <<<<<<<< end: rknn::RKNNMultiSurfacePass
D RKNN: [13:48:42.613] >>>>>> start: rknn::RKNNReplaceConstantTensorPass
D RKNN: [13:48:42.614] <<<<<<<< end: rknn::RKNNReplaceConstantTensorPass
D RKNN: [13:48:42.614] >>>>>> start: rknn::RKNNTilingPass
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:48:42.614] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
W RKNN: [13:48:42.615] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:48:42.615] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:48:42.615] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:48:42.615] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:48:42.615] Please help report this bug!
W RKNN: [13:48:42.615] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:48:42.615] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:48:42.615] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:48:42.615] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:48:42.615] Please help report this bug!
W RKNN: [13:48:42.615] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:48:42.615] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:48:42.615] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:48:42.615] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:48:42.615] Please help report this bug!
W RKNN: [13:48:42.615] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:48:42.615] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:48:42.615] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:48:42.615] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:48:42.615] Please help report this bug!
W RKNN: [13:48:42.615] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:48:42.615] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:48:42.615] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:48:42.615] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:48:42.615] Please help report this bug!
D RKNN: [13:48:42.615] <<<<<<<< end: rknn::RKNNTilingPass
D RKNN: [13:48:42.615] >>>>>> start: rknn::RKNNSubgraphManager
D RKNN: [13:48:42.615] <<<<<<<< end: rknn::RKNNSubgraphManager
D RKNN: [13:48:42.615] >>>>>> start: OpEmit
E RKNN: [13:48:42.615] REGTASK: The bit width of field value exceeds the limit, target: v2, offset: 0x500c, shift = 0, limit: 0x1fff, value: 0x20cf
E RKNN: [13:48:42.615] REGTASK: The bit width of field value exceeds the limit, target: v2, offset: 0x500c, shift = 0, limit: 0x1fff, value: 0x20cf
W RKNN: [13:48:42.615] Transpose will fallback to CPU, because input shape has exceeded the max limit, height(4) * width(8400) = 33600, required product no larger than 16384!
W RKNN: [13:48:42.615] Transpose will fallback to CPU, because input shape has exceeded the max limit, height(16) * width(8400) = 134400, required product no larger than 16384!
D RKNN: [13:48:42.615] <<<<<<<< end: OpEmit
D RKNN: [13:48:42.615] >>>>>> start: rknn::RKNNLayoutMatchPass
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.0/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.1/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/cv1/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/Split_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/Split_output_1]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/m.0/Add_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/Concat_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.2/cv2/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.3/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/cv1/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/Split_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/Split_output_1]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/m.0/Add_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/m.1/cv1/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/m.1/cv2/act/Mul_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/m.1/Add_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.4/Concat_output_0]
I RKNN: [13:48:42.615] AppointLayout: t->setNativeLayout(64), tname:[/model.5/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/Split_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/Split_output_1]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/m.0/Add_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/m.1/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/m.1/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/m.1/Add_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.6/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.7/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/Split_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/Split_output_1]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/m.0/Add_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.8/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.9/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.9/m/MaxPool_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.9/m_1/MaxPool_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.9/m_2/MaxPool_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.9/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.12/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.12/Split_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.12/Split_output_1]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.12/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.12/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.12/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.15/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.15/Split_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.15/Split_output_1]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.15/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.15/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.15/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.15/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.0/cv2.0.0/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.0/cv3.0.0/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.18/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.18/Split_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.18/Split_output_1]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.0/cv2.0.1/act/Mul_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.0/cv2.0.2/Conv_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.0/cv3.0.1/act/Mul_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.0/cv3.0.2/Conv_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Reshape_output_0_shape4_/model.22/Concat_3]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.18/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.18/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.18/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.18/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.1/cv2.1.0/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.1/cv3.1.0/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.21/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.21/Split_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.21/Split_output_1]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.1/cv2.1.1/act/Mul_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.1/cv2.1.2/Conv_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.1/cv3.1.1/act/Mul_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.1/cv3.1.2/Conv_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Concat_1_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Reshape_1_output_0_shape4_/model.22/Concat_3]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.21/m.0/cv1/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.21/m.0/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.21/Concat_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.21/cv2/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.2/cv2.2.0/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.2/cv3.2.0/act/Mul_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.2/cv2.2.1/act/Mul_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv2.2/cv2.2.2/Conv_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.2/cv3.2.1/act/Mul_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/cv3.2/cv3.2.2/Conv_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Concat_2_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Reshape_2_output_0_shape4_/model.22/Concat_3]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Concat_3_output_0]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/dfl/Softmax_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Slice_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Slice_1_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Sub_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Add_1_output_0]
W RKNN: [13:48:42.616] BroadcastLayoutConnFactory: op(Mul:/model.22/Mul_2) has undefined broadcast type, please check this op or contact rk-compiler-developer.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Sub_1_output_0]
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Add_2_output_0]
W RKNN: [13:48:42.616] BroadcastLayoutConnFactory: op(Mul:/model.22/Mul_2) has undefined broadcast type, please check this op or contact rk-compiler-developer.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Div_1_output_0]
W RKNN: [13:48:42.616] BroadcastLayoutConnFactory: op(Mul:/model.22/Mul_2) has undefined broadcast type, please check this op or contact rk-compiler-developer.
I RKNN: [13:48:42.616] AppointLayout: t->setNativeLayout(64), tname:[/model.22/Concat_4_output_0_shape4_tp_4/model.22/Concat_4]
W RKNN: [13:48:42.616] LayoutMatchManager: recursion_depth=3, Logic is Dangerous, Will Force layout to native.
D RKNN: [13:48:42.616] <<<<<<<< end: rknn::RKNNLayoutMatchPass
D RKNN: [13:48:42.616] >>>>>> start: rknn::RKNNAddSecondaryNode
D RKNN: [13:48:42.616] <<<<<<<< end: rknn::RKNNAddSecondaryNode
D RKNN: [13:48:42.616] >>>>>> start: OpEmit
D RKNN: [13:48:42.635] finish initComputeZoneMapByStepsVector
D RKNN: [13:48:42.635] finish initComputeZoneMapByStepsVector
D RKNN: [13:48:42.635] finish initComputeZoneMapByStepsVector
D RKNN: [13:48:42.635] finish initComputeZoneMapByStepsVector
D RKNN: [13:48:42.635] not need tranpose
D RKNN: [13:48:42.635] not need tranpose
D RKNN: [13:48:42.635] finish initComputeZoneMap
D RKNN: [13:48:42.635] emit max
E RKNN: [13:48:42.636] failed to config argb mode layer!
W __init__: rknn-toolkit2 version: 1.6.0+81f21f4d
--> Configuring model yolov8_small.onnx
done
--> Loading model
W load_onnx: It is recommended onnx opset 19, but your onnx model opset is 17!
W load_onnx: Model converted from pytorch, 'opset_version' should be set 19 in torch.onnx.export for successful convert!
Loading : 100%|███████████████████████████████████████████████| 143/143 [00:00<00:00, 184435.88it/s]
done
--> Building model
I base_optimize ...
I base_optimize done.
I
I fold_constant ...
I fold_constant done.
I
I correct_ops ...
I correct_ops done.
I
I fuse_ops ...
I fuse_ops results:
I replace_exswish: remove node = ['/model.0/act/Sigmoid', '/model.0/act/Mul'], add node = ['/model.0/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.1/act/Sigmoid', '/model.1/act/Mul'], add node = ['/model.1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.2/cv1/act/Sigmoid', '/model.2/cv1/act/Mul'], add node = ['/model.2/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.2/m.0/cv1/act/Sigmoid', '/model.2/m.0/cv1/act/Mul'], add node = ['/model.2/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.2/m.0/cv2/act/Sigmoid', '/model.2/m.0/cv2/act/Mul'], add node = ['/model.2/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.2/cv2/act/Sigmoid', '/model.2/cv2/act/Mul'], add node = ['/model.2/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.3/act/Sigmoid', '/model.3/act/Mul'], add node = ['/model.3/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.4/cv1/act/Sigmoid', '/model.4/cv1/act/Mul'], add node = ['/model.4/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.4/m.0/cv1/act/Sigmoid', '/model.4/m.0/cv1/act/Mul'], add node = ['/model.4/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.4/m.0/cv2/act/Sigmoid', '/model.4/m.0/cv2/act/Mul'], add node = ['/model.4/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.4/m.1/cv1/act/Sigmoid', '/model.4/m.1/cv1/act/Mul'], add node = ['/model.4/m.1/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.4/m.1/cv2/act/Sigmoid', '/model.4/m.1/cv2/act/Mul'], add node = ['/model.4/m.1/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.4/cv2/act/Sigmoid', '/model.4/cv2/act/Mul'], add node = ['/model.4/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.5/act/Sigmoid', '/model.5/act/Mul'], add node = ['/model.5/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.6/cv1/act/Sigmoid', '/model.6/cv1/act/Mul'], add node = ['/model.6/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.6/m.0/cv1/act/Sigmoid', '/model.6/m.0/cv1/act/Mul'], add node = ['/model.6/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.6/m.0/cv2/act/Sigmoid', '/model.6/m.0/cv2/act/Mul'], add node = ['/model.6/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.6/m.1/cv1/act/Sigmoid', '/model.6/m.1/cv1/act/Mul'], add node = ['/model.6/m.1/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.6/m.1/cv2/act/Sigmoid', '/model.6/m.1/cv2/act/Mul'], add node = ['/model.6/m.1/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.6/cv2/act/Sigmoid', '/model.6/cv2/act/Mul'], add node = ['/model.6/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.7/act/Sigmoid', '/model.7/act/Mul'], add node = ['/model.7/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.8/cv1/act/Sigmoid', '/model.8/cv1/act/Mul'], add node = ['/model.8/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.8/m.0/cv1/act/Sigmoid', '/model.8/m.0/cv1/act/Mul'], add node = ['/model.8/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.8/m.0/cv2/act/Sigmoid', '/model.8/m.0/cv2/act/Mul'], add node = ['/model.8/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.8/cv2/act/Sigmoid', '/model.8/cv2/act/Mul'], add node = ['/model.8/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.9/cv1/act/Sigmoid', '/model.9/cv1/act/Mul'], add node = ['/model.9/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.9/cv2/act/Sigmoid', '/model.9/cv2/act/Mul'], add node = ['/model.9/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.12/cv1/act/Sigmoid', '/model.12/cv1/act/Mul'], add node = ['/model.12/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.12/m.0/cv1/act/Sigmoid', '/model.12/m.0/cv1/act/Mul'], add node = ['/model.12/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.12/m.0/cv2/act/Sigmoid', '/model.12/m.0/cv2/act/Mul'], add node = ['/model.12/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.12/cv2/act/Sigmoid', '/model.12/cv2/act/Mul'], add node = ['/model.12/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.15/cv1/act/Sigmoid', '/model.15/cv1/act/Mul'], add node = ['/model.15/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.15/m.0/cv1/act/Sigmoid', '/model.15/m.0/cv1/act/Mul'], add node = ['/model.15/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.15/m.0/cv2/act/Sigmoid', '/model.15/m.0/cv2/act/Mul'], add node = ['/model.15/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.15/cv2/act/Sigmoid', '/model.15/cv2/act/Mul'], add node = ['/model.15/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.16/act/Sigmoid', '/model.16/act/Mul'], add node = ['/model.16/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.18/cv1/act/Sigmoid', '/model.18/cv1/act/Mul'], add node = ['/model.18/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.18/m.0/cv1/act/Sigmoid', '/model.18/m.0/cv1/act/Mul'], add node = ['/model.18/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.18/m.0/cv2/act/Sigmoid', '/model.18/m.0/cv2/act/Mul'], add node = ['/model.18/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.18/cv2/act/Sigmoid', '/model.18/cv2/act/Mul'], add node = ['/model.18/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.19/act/Sigmoid', '/model.19/act/Mul'], add node = ['/model.19/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.21/cv1/act/Sigmoid', '/model.21/cv1/act/Mul'], add node = ['/model.21/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.21/m.0/cv1/act/Sigmoid', '/model.21/m.0/cv1/act/Mul'], add node = ['/model.21/m.0/cv1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.21/m.0/cv2/act/Sigmoid', '/model.21/m.0/cv2/act/Mul'], add node = ['/model.21/m.0/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.21/cv2/act/Sigmoid', '/model.21/cv2/act/Mul'], add node = ['/model.21/cv2/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv2.2/cv2.2.0/act/Sigmoid', '/model.22/cv2.2/cv2.2.0/act/Mul'], add node = ['/model.22/cv2.2/cv2.2.0/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv2.2/cv2.2.1/act/Sigmoid', '/model.22/cv2.2/cv2.2.1/act/Mul'], add node = ['/model.22/cv2.2/cv2.2.1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv3.2/cv3.2.0/act/Sigmoid', '/model.22/cv3.2/cv3.2.0/act/Mul'], add node = ['/model.22/cv3.2/cv3.2.0/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv3.2/cv3.2.1/act/Sigmoid', '/model.22/cv3.2/cv3.2.1/act/Mul'], add node = ['/model.22/cv3.2/cv3.2.1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv2.1/cv2.1.0/act/Sigmoid', '/model.22/cv2.1/cv2.1.0/act/Mul'], add node = ['/model.22/cv2.1/cv2.1.0/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv2.1/cv2.1.1/act/Sigmoid', '/model.22/cv2.1/cv2.1.1/act/Mul'], add node = ['/model.22/cv2.1/cv2.1.1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv3.1/cv3.1.0/act/Sigmoid', '/model.22/cv3.1/cv3.1.0/act/Mul'], add node = ['/model.22/cv3.1/cv3.1.0/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv3.1/cv3.1.1/act/Sigmoid', '/model.22/cv3.1/cv3.1.1/act/Mul'], add node = ['/model.22/cv3.1/cv3.1.1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv2.0/cv2.0.0/act/Sigmoid', '/model.22/cv2.0/cv2.0.0/act/Mul'], add node = ['/model.22/cv2.0/cv2.0.0/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv2.0/cv2.0.1/act/Sigmoid', '/model.22/cv2.0/cv2.0.1/act/Mul'], add node = ['/model.22/cv2.0/cv2.0.1/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv3.0/cv3.0.0/act/Sigmoid', '/model.22/cv3.0/cv3.0.0/act/Mul'], add node = ['/model.22/cv3.0/cv3.0.0/act/Sigmoid_2swish']
I replace_exswish: remove node = ['/model.22/cv3.0/cv3.0.1/act/Sigmoid', '/model.22/cv3.0/cv3.0.1/act/Mul'], add node = ['/model.22/cv3.0/cv3.0.1/act/Sigmoid_2swish']
I replace_parallel_slice_by_split: remove node = ['/model.22/Slice', '/model.22/Slice_1'], add node = ['/model.22/Slice_2split']
I unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Concat_3_0_unsqueeze0', '/model.22/Concat_3_1_unsqueeze0', '/model.22/Concat_3_2_unsqueeze0', '/model.22/Concat_3_0_unsqueeze1']
I unsqueeze_to_4d_split: remove node = [], add node = ['/model.22/Split_0_unsqueeze0', '/model.22/Split_0_unsqueeze1', '/model.22/Split_1_unsqueeze1']
I convert_softmax_to_exsoftmax13: remove node = ['/model.22/dfl/Softmax'], add node = ['/model.22/dfl/Softmax']
I unsqueeze_to_4d_split: remove node = [], add node = ['/model.22/Slice_2split_0_unsqueeze0', '/model.22/Slice_2split_0_unsqueeze1', '/model.22/Slice_2split_1_unsqueeze1']
I unsqueeze_to_4d_sub: remove node = [], add node = ['/model.22/Sub_1_unsqueeze0', '/model.22/Sub_0_unsqueeze1']
I unsqueeze_to_4d_add: remove node = [], add node = ['/model.22/Add_1_1_unsqueeze0', '/model.22/Add_1_0_unsqueeze1']
I unsqueeze_to_4d_add: remove node = [], add node = ['/model.22/Add_2_0_unsqueeze0', '/model.22/Add_2_1_unsqueeze0', '/model.22/Add_2_0_unsqueeze1']
I convert_div_to_mul: remove node = ['/model.22/Div_1'], add node = ['/model.22/Div_1_2mul']
I unsqueeze_to_4d_sub: remove node = [], add node = ['/model.22/Sub_1_0_unsqueeze0', '/model.22/Sub_1_1_unsqueeze0', '/model.22/Sub_1_0_unsqueeze1']
I unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Concat_4_0_unsqueeze0', '/model.22/Concat_4_1_unsqueeze0', '/model.22/Concat_4_0_unsqueeze1']
I unsqueeze_to_4d_mul: remove node = [], add node = ['/model.22/Mul_2_0_unsqueeze0', '/model.22/Mul_2_0_unsqueeze1']
I unsqueeze_to_4d_sigmoid: remove node = [], add node = ['/model.22/Sigmoid_0_unsqueeze0', '/model.22/Sigmoid_0_unsqueeze1']
I unsqueeze_to_4d_concat: remove node = [], add node = ['/model.22/Concat_5_0_unsqueeze0', '/model.22/Concat_5_1_unsqueeze0', '/model.22/Concat_5_0_unsqueeze1']
I fuse_two_reshape: remove node = ['/model.22/Reshape_2', '/model.22/Reshape_1', '/model.22/Reshape', '/model.22/Concat_3_0_unsqueeze1', '/model.22/Split_0_unsqueeze1', '/model.22/dfl/Reshape_1']
I convert_split_to_conv_split: remove node = [], add node = ['/model.22/Slice_output_0_shape4_conv', '/model.22/dfl/Reshape_1_output_0_shape4_/model.22/Slice_2split_conv_/model.22/Slice_2split']
I fuse_two_reshape: remove node = ['/model.22/Slice_2split_0_unsqueeze1']
I remove_parallel_reshape: remove node = ['/model.22/Add_2_0_unsqueeze0']
I fuse_two_reshape: remove node = ['/model.22/Slice_2split_1_unsqueeze1']
I remove_parallel_reshape: remove node = ['/model.22/Add_2_1_unsqueeze0']
I unsqueeze_to_4d_mul: remove node = [], add node = ['/model.22/Div_1_2mul_0_unsqueeze0', '/model.22/Div_1_2mul_0_unsqueeze1']
I swap_concat_axis_avoid_channel_concat: remove node = [], add node = ['/model.22/Concat_4_swap_concat_reshape_i0_out', '/model.22/Concat_4_swap_concat_reshape_i1_out', '/model.22/Concat_4_swap_concat_reshape_o0_out']
I fuse_two_reshape: remove node = ['/model.22/Concat_4_0_unsqueeze1', '/model.22/Mul_2_0_unsqueeze1', '/model.22/Split_1_unsqueeze1', '/model.22/Sigmoid_0_unsqueeze1']
I swap_concat_axis_avoid_channel_concat: remove node = [], add node = ['/model.22/Concat_5_swap_concat_reshape_i0_out', '/model.22/Concat_5_swap_concat_reshape_i1_out', '/model.22/Concat_5_swap_concat_reshape_o0_out']
I remove_invalid_reshape: remove node = ['/model.22/Split_0_unsqueeze0', '/model.22/Sub_1_unsqueeze0']
I fuse_two_reshape: remove node = ['/model.22/Sub_0_unsqueeze1']
I remove_invalid_reshape: remove node = ['/model.22/Add_1_1_unsqueeze0']
I fuse_two_reshape: remove node = ['/model.22/Add_1_0_unsqueeze1', '/model.22/Sub_1_0_unsqueeze1', '/model.22/Add_2_0_unsqueeze1', '/model.22/Div_1_2mul_0_unsqueeze1', '/model.22/Concat_4_swap_concat_reshape_o0_out', '/model.22/Concat_5_0_unsqueeze0']
I remove_invalid_reshape: remove node = ['/model.22/Sigmoid_0_unsqueeze0']
I fuse_two_reshape: remove node = ['/model.22/Concat_5_1_unsqueeze0', '/model.22/Concat_5_swap_concat_reshape_o0_out']
I remove_invalid_reshape: remove node = ['/model.22/Sub_1_1_unsqueeze0', '/model.22/Sub_1_0_unsqueeze0']
I fuse_two_reshape: remove node = ['/model.22/Concat_4_1_unsqueeze0']
I remove_invalid_reshape: remove node = ['/model.22/Div_1_2mul_0_unsqueeze0']
I fuse_two_reshape: remove node = ['/model.22/Concat_4_0_unsqueeze0']
I fuse_reshape_into_conv: remove node = ['/model.22/Slice_2split_0_unsqueeze0', '/model.22/dfl/Reshape_1_output_0_shape4_/model.22/Slice_2split_conv_/model.22/Slice_2split'], add node = ['/model.22/dfl/Reshape_1_output_0_shape4_/model.22/Slice_2split_conv_/model.22/Slice_2split']
I remove_reshape_after_concat: remove node = ['/model.22/Mul_2_0_unsqueeze0']
I remove_invalid_reshape: remove node = ['/model.22/Concat_4_swap_concat_reshape_i1_out', '/model.22/Concat_4_swap_concat_reshape_i0_out']
I convert_concat_to_conv_concat: remove node = [], add node = ['/model.22/Div_1_output_0_conv_/model.22/Concat_4', '/model.22/Concat_4_output_0_shape4_/model.22/Mul_2_conv']
I fuse_mul_into_conv2: remove node = ['/model.22/Div_1_2mul']
I fold_constant ...
I fold_constant done.
I fuse_ops done.
I
W build: found outlier value, this may affect quantization accuracy
const name abs_mean abs_std outlier value
model.0.conv.weight 4.03 4.41 26.039
I sparse_weight ...
I sparse_weight done.
I
GraphPreparing : 100%|██████████████████████████████████████████| 180/180 [00:00<00:00, 2094.71it/s]
Quantizating : 100%|█████████████████████████████████████████████| 180/180 [00:00<00:00, 317.54it/s]
I
I quant_optimizer ...
I quant_optimizer results:
I adjust_tanh_sigmoid: ['/model.22/Sigmoid']
I adjust_concat_split: ['/model.22/Split', '/model.22/Concat_3', '/model.22/Concat', '/model.22/Concat_1', '/model.22/Concat_2']
I adjust_no_change_node: ['/model.22/Concat_3_0_unsqueeze0', '/model.22/Concat_3_1_unsqueeze0', '/model.22/Concat_3_2_unsqueeze0', '/model.9/m_2/MaxPool', '/model.9/m_1/MaxPool', '/model.9/m/MaxPool']
I quant_optimizer done.
I
I recover_const_share ...
I recover_const_share done.
I
W build: The default input dtype of 'images' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
W build: The default output dtype of 'output0' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
I rknn building ...
I RKNN: [13:55:01.955] compress = 0, conv_eltwise_activation_fuse = 1, global_fuse = 1, multi-core-model-mode = 7, output_optimize = 1,enable_argb_group=0 ,layout_match = 0, pipeline_fuse = 0
I RKNN: librknnc version: 1.6.0 (585b3edcf@2023-12-11T08:03:14)
D RKNN: [13:55:01.002] RKNN is invoked
W RKNN: [13:55:02.191] Model initializer tensor data is empty, name: empty_placeholder_0
D RKNN: [13:55:02.194] >>>>>> start: rknn::RKNNExtractCustomOpAttrs
D RKNN: [13:55:02.194] <<<<<<<< end: rknn::RKNNExtractCustomOpAttrs
D RKNN: [13:55:02.194] >>>>>> start: rknn::RKNNSetOpTargetPass
D RKNN: [13:55:02.194] <<<<<<<< end: rknn::RKNNSetOpTargetPass
D RKNN: [13:55:02.194] >>>>>> start: rknn::RKNNBindNorm
D RKNN: [13:55:02.194] <<<<<<<< end: rknn::RKNNBindNorm
D RKNN: [13:55:02.194] >>>>>> start: rknn::RKNNAddFirstConv
D RKNN: [13:55:02.194] <<<<<<<< end: rknn::RKNNAddFirstConv
D RKNN: [13:55:02.194] >>>>>> start: rknn::RKNNEliminateQATDataConvert
D RKNN: [13:55:02.195] <<<<<<<< end: rknn::RKNNEliminateQATDataConvert
D RKNN: [13:55:02.195] >>>>>> start: rknn::RKNNTileGroupConv
D RKNN: [13:55:02.195] <<<<<<<< end: rknn::RKNNTileGroupConv
D RKNN: [13:55:02.195] >>>>>> start: rknn::RKNNAddConvBias
D RKNN: [13:55:02.195] <<<<<<<< end: rknn::RKNNAddConvBias
D RKNN: [13:55:02.195] >>>>>> start: rknn::RKNNTileChannel
D RKNN: [13:55:02.195] <<<<<<<< end: rknn::RKNNTileChannel
D RKNN: [13:55:02.195] >>>>>> start: rknn::RKNNPerChannelPrep
D RKNN: [13:55:02.195] <<<<<<<< end: rknn::RKNNPerChannelPrep
D RKNN: [13:55:02.195] >>>>>> start: rknn::RKNNBnQuant
D RKNN: [13:55:02.195] <<<<<<<< end: rknn::RKNNBnQuant
D RKNN: [13:55:02.195] >>>>>> start: rknn::RKNNFuseOptimizerPass
D RKNN: [13:55:02.209] <<<<<<<< end: rknn::RKNNFuseOptimizerPass
D RKNN: [13:55:02.209] >>>>>> start: rknn::RKNNTurnAutoPad
D RKNN: [13:55:02.209] <<<<<<<< end: rknn::RKNNTurnAutoPad
D RKNN: [13:55:02.209] >>>>>> start: rknn::RKNNInitRNNConst
D RKNN: [13:55:02.209] <<<<<<<< end: rknn::RKNNInitRNNConst
D RKNN: [13:55:02.209] >>>>>> start: rknn::RKNNInitCastConst
D RKNN: [13:55:02.209] <<<<<<<< end: rknn::RKNNInitCastConst
D RKNN: [13:55:02.209] >>>>>> start: rknn::RKNNMultiSurfacePass
D RKNN: [13:55:02.209] <<<<<<<< end: rknn::RKNNMultiSurfacePass
D RKNN: [13:55:02.209] >>>>>> start: rknn::RKNNReplaceConstantTensorPass
D RKNN: [13:55:02.210] <<<<<<<< end: rknn::RKNNReplaceConstantTensorPass
D RKNN: [13:55:02.210] >>>>>> start: rknn::RKNNTilingPass
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.210] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
W RKNN: [13:55:02.211] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:55:02.211] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:55:02.211] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:55:02.211] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:55:02.211] Please help report this bug!
W RKNN: [13:55:02.211] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:55:02.211] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:55:02.211] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:55:02.211] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:55:02.211] Please help report this bug!
W RKNN: [13:55:02.211] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:55:02.211] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:55:02.211] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:55:02.211] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:55:02.211] Please help report this bug!
W RKNN: [13:55:02.211] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:55:02.211] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:55:02.211] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:55:02.211] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:55:02.211] Please help report this bug!
W RKNN: [13:55:02.211] Failed to config layer: 'Conv:/model.22/dfl/conv/Conv' using 3Core fallback to single core mode,
W RKNN: [13:55:02.211] core_num 3 ori_Ih 4 ori_Iw 8400 ori_Ic 16 ori_Ib 1
W RKNN: [13:55:02.211] ori_Kh 1 ori_Kw 1 ori_Kk 1 ori_Kc 16 ori_Ksx 1 ori_Ksy 1
W RKNN: [13:55:02.211] ori_Oh 4 oriOw 8400 oriOc 1 pad_t 0 pad_b 0 pad_l 0 pad_r 0,
W RKNN: [13:55:02.211] Please help report this bug!
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 2100, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] DatainEntries overflow, need to tiling more, datain_entries: 4200, limitation: 2048
D RKNN: [13:55:02.211] <<<<<<<< end: rknn::RKNNTilingPass
D RKNN: [13:55:02.211] >>>>>> start: OpEmit
D RKNN: [13:55:02.211] Transpose will fallback to CPU, because input shape has exceeded the max limit, height(16) * width(8400) = 134400, required product no larger than 16384!
D RKNN: [13:55:02.212] <<<<<<<< end: OpEmit
D RKNN: [13:55:02.212] >>>>>> start: rknn::RKNNLayoutMatchPass
D RKNN: [13:55:02.212] <<<<<<<< end: rknn::RKNNLayoutMatchPass
D RKNN: [13:55:02.212] >>>>>> start: rknn::RKNNAddSecondaryNode
D RKNN: [13:55:02.212] <<<<<<<< end: rknn::RKNNAddSecondaryNode
D RKNN: [13:55:02.212] >>>>>> start: OpEmit
D RKNN: [13:55:02.228] Transpose will fallback to CPU, because input shape has exceeded the max limit, height(16) * width(8400) = 134400, required product no larger than 16384!
D RKNN: [13:55:02.228] not need tranpose
D RKNN: [13:55:02.228] not need tranpose
D RKNN: [13:55:02.228] finish initComputeZoneMap
D RKNN: [13:55:02.228] emit max
E RKNN: [13:55:02.228] failed to config argb mode layer!
None of the versions can convert this model. |
RKNN-Toolkit 2.2.0 seems OK. The reason all the outputs were 0 is caused by quantization. When you concat the score(range 0->1) and position value(range 0->height/width), the score value will be suppressed to 0. Two ways to solve this:
And there is already a yolov8 demo in this repo: https://github.com/airockchip/rknn_model_zoo |
I'm trying to compile the model yolov8 small for RK3588
model_on_google_disk
rknn-toolkit2 2.2.0
Compile code simple:
The detailed log output shows some errors.
After running the compilation I get the model. However, when I try to run the inference on the RK3588, I get a vector of zero length
From the detailed output of the model compilation I believe that the model was compiled incorrectly.
Can you help with the solution?
The text was updated successfully, but these errors were encountered: