- 模型转换的配置文件
model_parameters:
# 当前推荐使用 CPU postprocess 导出图:
# segmentation_background_lowres : 3 通道，背景头 logits，90x160
# segmentation_foreground_lowres : 13 通道，前景头 logits，90x160
# depth : 1 通道，保持 360x640
# 其中双头 segmentation 会在本地 / 板端脚本里做 bilinear upsample + merge。
# 这样可以把 /model/Resize 和 /model/Resize_1 从量化主干里移出去，
# 优先规避当前最敏感的 segmentation head 上采样量化路径。
onnx_model: '../lraspp_99_torch110.onnx'
march: 'bayes-e'
output_model_file_prefix: 'lraspp_99_torch110_2'
working_dir: '../hb_mapper_output_2/lraspp_99_torch110_2'
layer_out_dump: False
node_info: {"711_Cast": {'ON': 'BPU','InputType': 'int16','OutputType': 'int16'},
"Where_263_cast": {'ON': 'BPU','InputType': 'int16','OutputType': 'int16'},
"Where_263_equal": {'ON': 'BPU','InputType': 'int8','OutputType': 'int8'},
"Where_263_equal_output_cast": {'ON': 'BPU','InputType': 'int8','OutputType': 'int8'},
"ScatterND_337": {'ON': 'BPU','InputType': 'int16','OutputType': 'int16'},
"Where_340_cast": {'ON': 'BPU','InputType': 'int16','OutputType': 'int16'},
"Where_340_equal": {'ON': 'BPU','InputType': 'int16','OutputType': 'int16'},
"Where_340_equal_output_cast": {'ON': 'BPU','InputType': 'int16','OutputType': 'int16'},
"ScatterND_414": {'ON': 'BPU','InputType': 'int16','OutputType': 'int16'},
}
input_parameters:
input_name: 'input'
# 当前最新模型真实输入不是纯 RGB，而是 5 通道 featuremap:
# 0~2: RGB，已做 (x/255 - mean) / std
# 3 : radar_range，先按 sensor_depth_max 解码到米制，再按 [depth_min, depth_max]
# 截断并线性归一化到 [0, 1]
# 4 : radar_valid_mask，取值 {0, 1}
# 该输入口径必须与 export_onnx.py、test_and_visualize_onnx.py、
# test_and_visualize_RDK_X5.py 保持完全一致。
input_type_train: 'featuremap'
input_layout_train: 'NCHW'
input_type_rt: 'featuremap'
input_layout_rt: 'NCHW'
input_shape: '1x5x360x640'
input_batch: 1
# featuremap 输入要求上板前已经完成全部前处理，因此这里不能再走 rgb/nv12 预处理链
norm_type: 'no_preprocess'
calibration_parameters:
# 正式 PTQ 时，这里需要放“预处理后的 5 通道 float32 featuremap 校准数据”
# 数据口径必须与 input_type_train/input_layout_train/input_shape 完全一致
# 即每个样本都是 row-major 的 float32 二进制，shape = 1x5x360x640 或 5x360x640
# 当前 featuremap 生成逻辑应与 02_generate_calibration_data.py 的最新版本一致。
cal_data_dir: '../hb_calibration_data_f32'
cal_data_type: 'float32'
preprocess_on: False
# per_channel: True
# 如果已经准备好了代表性的最新 featuremap 校准集，建议保持 default 做正式 PTQ
# calibration_type: 'default'
calibration_type: 'skip'
compiler_parameters:
compile_mode: 'latency'
debug: True
core_num: 1
optimize_level: 'O3'
# featuremap 输入通常从 DDR 喂入
input_source: {'input': 'ddr'}
- hb_mapper_makertbin.log日志
2026-04-20 11:01:26,668 file: tool_utils.py func: tool_utils line No: 77 log will be stored in /open_explorer/test_lraspp_model_0420/hb_mapper_makertbin.log
2026-04-20 11:01:26,668 file: hb_mapper.py func: hb_mapper line No: 132 Start hb_mapper....
2026-04-20 11:01:26,668 file: hb_mapper.py func: hb_mapper line No: 133 hbdk version 3.49.15
2026-04-20 11:01:26,668 file: hb_mapper.py func: hb_mapper line No: 134 horizon_nn version 1.1.0
2026-04-20 11:01:26,668 file: hb_mapper.py func: hb_mapper line No: 135 hb_mapper version 1.24.3
2026-04-20 11:01:26,668 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 530 Start Model Convert....
2026-04-20 11:01:26,674 file: mapper_conf_parser.py func: mapper_conf_parser line No: 105 validating model_parameters...
2026-04-20 11:01:26,674 file: mapper_conf_parser.py func: mapper_conf_parser line No: 1347 Using abs path /open_explorer/test_lraspp_model_0420/lraspp_99_torch110.onnx
2026-04-20 11:01:26,674 file: mapper_conf_parser.py func: mapper_conf_parser line No: 260 Using onnx model file: /open_explorer/test_lraspp_model_0420/lraspp_99_torch110.onnx
2026-04-20 11:01:26,684 file: onnx_parser.py func: onnx_parser line No: 39 Model input names: ['input']
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 264 Model has 1 inputs according to model file
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 1347 Using abs path /open_explorer/test_lraspp_model_0420/hb_mapper_output/lraspp_99_torch110
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 287 working_dir does not exist. Creating working_dir: /open_explorer/test_lraspp_model_0420/hb_mapper_output/lraspp_99_torch110
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 438 node_dict: {self.node_dict}
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 119 validating model_parameters finished
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 123 validating input_parameters...
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 135 validating input_parameters finished
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 139 validating calibration_parameters...
2026-04-20 11:01:26,684 file: mapper_conf_parser.py func: mapper_conf_parser line No: 1347 Using abs path /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 1013 The calibration dir name suffix is the same as the value float32 of the cal_data_type parameter and will be read with the value of cal_data_type.
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 155 validating calibration_parameters finished
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 159 validating custom_op...
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 1076 custom_op does not exist, skipped
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 165 validating custom_op finished
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 168 validating compiler_parameters...
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 183 validating compiler_parameters finished
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 187 validating deprecated parameters...
2026-04-20 11:01:26,685 file: mapper_conf_parser.py func: mapper_conf_parser line No: 193 validating deprecated parameters finished
2026-04-20 11:01:26,685 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 54 Dump config:
2026-04-20 11:01:26,685 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 55 calibration_parameters:
cal_data_dir: ../hb_calibration_data_f32
cal_data_type: float32
calibration_type: default
preprocess_on: false
compiler_parameters:
compile_mode: latency
core_num: 1
debug: true
input_source:
input: ddr
optimize_level: O3
input_parameters:
input_batch: '1'
input_layout_rt: NCHW
input_layout_train: NCHW
input_name: input
input_shape: 1x5x360x640
input_type_rt: featuremap
input_type_train: featuremap
norm_type: no_preprocess
model_parameters:
layer_out_dump: false
march: bayes-e
node_info:
711_Cast:
InputType: int16
'ON': BPU
OutputType: int16
ScatterND_337:
InputType: int16
'ON': BPU
OutputType: int16
ScatterND_414:
InputType: int16
'ON': BPU
OutputType: int16
Where_263_cast:
InputType: int16
'ON': BPU
OutputType: int16
Where_263_equal:
InputType: int8
'ON': BPU
OutputType: int8
Where_263_equal_output_cast:
InputType: int8
'ON': BPU
OutputType: int8
Where_340_cast:
InputType: int16
'ON': BPU
OutputType: int16
Where_340_equal:
InputType: int16
'ON': BPU
OutputType: int16
Where_340_equal_output_cast:
InputType: int16
'ON': BPU
OutputType: int16
onnx_model: ../lraspp_99_torch110.onnx
output_model_file_prefix: lraspp_99_torch110
working_dir: ../hb_mapper_output/lraspp_99_torch110
2026-04-20 11:01:26,689 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 60 input 'input' : original model shape: [1, 5, 360, 640]
2026-04-20 11:01:26,689 file: loader.py func: loader line No: 204 *******************************************
2026-04-20 11:01:26,689 file: loader.py func: loader line No: 205 First calibration picture name: scene_003_27.feature
2026-04-20 11:01:26,689 file: loader.py func: loader line No: 207 First calibration picture md5:
2026-04-20 11:01:26,696 file: loader.py func: loader line No: 211 *******************************************
2026-04-20 11:01:26,697 file: loader.py func: loader line No: 282 created RawImageDirLoader of shape:[1, 5, 360, 640]
2026-04-20 11:01:26,697 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_003_27.feature
2026-04-20 11:01:26,723 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_004_25.feature
2026-04-20 11:01:26,730 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_006_39.feature
2026-04-20 11:01:26,739 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_007_40.feature
2026-04-20 11:01:26,743 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_008_10.feature
2026-04-20 11:01:26,746 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_009_54.feature
2026-04-20 11:01:26,748 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_011_00.feature
2026-04-20 11:01:26,752 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_012_02.feature
2026-04-20 11:01:26,755 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_013_16.feature
2026-04-20 11:01:26,757 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_015_16.feature
2026-04-20 11:01:26,759 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_018_28.feature
2026-04-20 11:01:26,762 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_019_42.feature
2026-04-20 11:01:26,764 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_021_41.feature
2026-04-20 11:01:26,767 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_022_13.feature
2026-04-20 11:01:26,769 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_023_27.feature
2026-04-20 11:01:26,771 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_025_20.feature
2026-04-20 11:01:26,773 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_026_26.feature
2026-04-20 11:01:26,775 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_027_54.feature
2026-04-20 11:01:26,777 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_028_40.feature
2026-04-20 11:01:26,779 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_029_58.feature
2026-04-20 11:01:26,781 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_033_12.feature
2026-04-20 11:01:26,805 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_034_22.feature
2026-04-20 11:01:26,807 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_038_03.feature
2026-04-20 11:01:26,809 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_041_51.feature
2026-04-20 11:01:26,811 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_042_52.feature
2026-04-20 11:01:26,813 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_043_12.feature
2026-04-20 11:01:26,815 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_044_32.feature
2026-04-20 11:01:26,817 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_045_32.feature
2026-04-20 11:01:26,819 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_046_47.feature
2026-04-20 11:01:26,820 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_047_42.feature
2026-04-20 11:01:26,822 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_048_20.feature
2026-04-20 11:01:26,824 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_101_18.feature
2026-04-20 11:01:26,826 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_102_11.feature
2026-04-20 11:01:26,829 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_103_35.feature
2026-04-20 11:01:26,831 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_104_38.feature
2026-04-20 11:01:26,833 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_105_58.feature
2026-04-20 11:01:26,835 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_107_47.feature
2026-04-20 11:01:26,837 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_108_45.feature
2026-04-20 11:01:26,853 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_109_46.feature
2026-04-20 11:01:26,855 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_111_40.feature
2026-04-20 11:01:26,857 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_112_29.feature
2026-04-20 11:01:26,859 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_113_21.feature
2026-04-20 11:01:26,861 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_114_27.feature
2026-04-20 11:01:26,864 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_115_53.feature
2026-04-20 11:01:26,866 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_116_20.feature
2026-04-20 11:01:26,868 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_117_45.feature
2026-04-20 11:01:26,870 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_118_19.feature
2026-04-20 11:01:26,872 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_119_28.feature
2026-04-20 11:01:26,874 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_122_02.feature
2026-04-20 11:01:26,876 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_123_53.feature
2026-04-20 11:01:26,878 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_124_13.feature
2026-04-20 11:01:26,880 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_125_09.feature
2026-04-20 11:01:26,883 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_126_01.feature
2026-04-20 11:01:26,884 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_127_37.feature
2026-04-20 11:01:26,887 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_128_52.feature
2026-04-20 11:01:26,889 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_129_11.feature
2026-04-20 11:01:26,901 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_131_26.feature
2026-04-20 11:01:26,904 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_132_32.feature
2026-04-20 11:01:26,906 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_133_42.feature
2026-04-20 11:01:26,908 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_134_53.feature
2026-04-20 11:01:26,910 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_135_32.feature
2026-04-20 11:01:26,912 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_136_49.feature
2026-04-20 11:01:26,914 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_137_37.feature
2026-04-20 11:01:26,916 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_139_45.feature
2026-04-20 11:01:26,918 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_141_41.feature
2026-04-20 11:01:26,920 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_143_39.feature
2026-04-20 11:01:26,922 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_144_37.feature
2026-04-20 11:01:26,924 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_145_30.feature
2026-04-20 11:01:26,926 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_146_32.feature
2026-04-20 11:01:26,929 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_147_26.feature
2026-04-20 11:01:26,931 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_148_47.feature
2026-04-20 11:01:26,933 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_211_23.feature
2026-04-20 11:01:26,935 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_212_30.feature
2026-04-20 11:01:26,937 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_213_30.feature
2026-04-20 11:01:26,952 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_214_47.feature
2026-04-20 11:01:26,955 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_216_30.feature
2026-04-20 11:01:26,957 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_217_37.feature
2026-04-20 11:01:26,959 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_218_02.feature
2026-04-20 11:01:26,961 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_219_18.feature
2026-04-20 11:01:26,963 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_221_48.feature
2026-04-20 11:01:26,965 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_222_55.feature
2026-04-20 11:01:26,967 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_224_55.feature
2026-04-20 11:01:26,969 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_225_17.feature
2026-04-20 11:01:26,971 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_226_14.feature
2026-04-20 11:01:26,974 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_227_10.feature
2026-04-20 11:01:26,976 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_232_01.feature
2026-04-20 11:01:26,978 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_233_35.feature
2026-04-20 11:01:26,980 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_234_28.feature
2026-04-20 11:01:26,982 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_235_46.feature
2026-04-20 11:01:26,985 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_236_58.feature
2026-04-20 11:01:26,987 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_237_40.feature
2026-04-20 11:01:26,998 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_238_35.feature
2026-04-20 11:01:27,000 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_242_40.feature
2026-04-20 11:01:27,002 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_243_58.feature
2026-04-20 11:01:27,005 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_244_56.feature
2026-04-20 11:01:27,007 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_245_16.feature
2026-04-20 11:01:27,009 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_246_43.feature
2026-04-20 11:01:27,011 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_247_46.feature
2026-04-20 11:01:27,013 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_248_31.feature
2026-04-20 11:01:27,016 file: loader.py func: loader line No: 287 Read raw file: /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32/scene_249_20.feature
2026-04-20 11:01:27,018 file: tool_utils.py func: tool_utils line No: 368 num of calibration data: 100
2026-04-20 11:01:27,018 file: tool_utils.py func: tool_utils line No: 369 calibration data shape: (1, 5, 360, 640)
2026-04-20 11:01:27,066 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 519 call build params:
{'march': 'bayes-e', 'save_model': True, 'name_prefix': 'lraspp_99_torch110', 'input_dict': {'input': {'input_shape': [1, 5, 360, 640], 'input_batch': 1, 'original_input_layout': 'NCHW'}}, 'hbdk_dict': {'hbdk_pass_through_params': '--O3 --debug --core-num 1 --fast ', 'input-source': {'input': 'ddr', '_default_value': 'ddr'}}, 'node_dict': {'711_Cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_263_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_263_equal': {'ON': 'BPU', 'InputType': 'int8', 'OutputType': 'int8'}, 'Where_263_equal_output_cast': {'ON': 'BPU', 'InputType': 'int8', 'OutputType': 'int8'}, 'ScatterND_337': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_equal': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_equal_output_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'ScatterND_414': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}}, 'check_mode': False}
2026-04-20 11:01:27,077 file: model_builder.py func: model_builder line No: 35 Start to Horizon NN Model Convert.
2026-04-20 11:01:27,093 file: model_debugger.py func: model_debugger line No: 67 Loading horizon_nn debug methods:set()
2026-04-20 11:01:27,094 file: quantization_config.py func: quantization_config line No: 305 The activation calibration parameters:
calibration_type: ['max', 'kl']
max_percentile: [0.99995, 1.0]
per_channel: [True, False]
asymmetric: [True, False]
The modelwise search parameters:
similarity: 0.995
metric: cosine-similarity
The input of node 711_Cast are set to : int16
The output of node 711_Cast are set to : int16
The input of node Where_263_cast are set to : int16
The output of node Where_263_cast are set to : int16
The input of node Where_263_equal are set to : int8
The output of node Where_263_equal are set to : int8
The input of node Where_263_equal_output_cast are set to : int8
The output of node Where_263_equal_output_cast are set to : int8
The input of node ScatterND_337 are set to : int16
The output of node ScatterND_337 are set to : int16
The input of node Where_340_cast are set to : int16
The output of node Where_340_cast are set to : int16
The input of node Where_340_equal are set to : int16
The output of node Where_340_equal are set to : int16
The input of node Where_340_equal_output_cast are set to : int16
The output of node Where_340_equal_output_cast are set to : int16
The input of node ScatterND_414 are set to : int16
The output of node ScatterND_414 are set to : int16
There are 9 nodes designated to run on the bpu: ['711_Cast', 'Where_263_cast', 'Where_263_equal', 'Where_263_equal_output_cast', 'ScatterND_337', 'Where_340_cast', 'Where_340_equal', 'Where_340_equal_output_cast', 'ScatterND_414'].
2026-04-20 11:01:27,094 file: model_builder.py func: model_builder line No: 197 The specified model compilation architecture: bayes-e.
2026-04-20 11:01:27,094 file: model_builder.py func: model_builder line No: 207 The specified model compilation optimization parameters: [].
2026-04-20 11:01:27,094 file: model_builder.py func: model_builder line No: 35 Start to prepare the onnx model.
2026-04-20 11:01:27,150 file: prepare.py func: prepare line No: 106 Input ONNX Model Information:
ONNX IR version: 7
Opset version: ['ai.onnx v11', 'horizon v1']
Producer: pytorch v1.10
Domain: None
Version: None
Graph input:
input: shape=[1, 5, 360, 640], dtype=FLOAT32
Graph output:
...entation: shape=['ScatterNDsegmentation_dim_0', 'ScatterNDsegmentation_dim_1', 'ScatterNDsegmentation_dim_2', 'ScatterNDsegmentation_dim_3'], dtype=FLOAT32
depth: shape=['Sigmoiddepth_dim_0', 1, 'Sigmoiddepth_dim_2', 'Sigmoiddepth_dim_3'], dtype=FLOAT32
2026-04-20 11:01:27,232 file: model_builder.py func: model_builder line No: 38 End to prepare the onnx model.
2026-04-20 11:01:27,256 file: model_builder.py func: model_builder line No: 265 Saving model to: lraspp_99_torch110_original_float_model.onnx.
2026-04-20 11:01:27,256 file: model_builder.py func: model_builder line No: 35 Start to optimize the onnx model.
2026-04-20 11:01:28,839 file: constant_folding.py func: constant_folding line No: 66 Summary info for constant_folding:
2026-04-20 11:01:28,839 file: constant_folding.py func: constant_folding line No: 67 After constant_folding, the number of nodes has changed from 415 to 207.
2026-04-20 11:01:28,839 file: constant_folding.py func: constant_folding line No: 71 After constant_folding, the number of parameters has changed from 3407002 to 22299713.
2026-04-20 11:01:28,839 file: constant_folding.py func: constant_folding line No: 76 Detailed info for constant_folding:
2026-04-20 11:01:28,839 file: constant_folding.py func: constant_folding line No: 88 After folding node (op_name: ConstantOfShape_244, op_type: ConstantOfShape), the number of increased parameters is 3455995.
After folding node (op_name: Expand_380, op_type: Expand), the number of increased parameters is 2995195.
After folding node (op_name: Expand_388, op_type: Expand), the number of increased parameters is 2995183.
After folding node (op_name: Expand_396, op_type: Expand), the number of increased parameters is 2994836.
After folding node (op_name: Expand_404, op_type: Expand), the number of increased parameters is 2994556.
After folding node (op_name: Add_372, op_type: Add), the number of increased parameters is 2990520.
After folding node (op_name: Gather_262, op_type: Gather), the number of increased parameters is 691197.
After folding node (op_name: Expand_303, op_type: Expand), the number of increased parameters is 691195.
After folding node (op_name: Expand_311, op_type: Expand), the number of increased parameters is 691193.
After folding node (op_name: Expand_319, op_type: Expand), the number of increased parameters is 690836.
After folding node (op_name: Expand_327, op_type: Expand), the number of increased parameters is 690556.
After folding node (op_name: Add_295, op_type: Add), the number of increased parameters is 690120.
After folding node (op_name: Add_371, op_type: Add), the number of increased parameters is 4667.
After folding node (op_name: Add_294, op_type: Add), the number of increased parameters is 1077.
After folding node (op_name: Range_286, op_type: Range), the number of increased parameters is 637.
After folding node (op_name: Range_363, op_type: Range), the number of increased parameters is 637.
After folding node (op_name: Range_279, op_type: Range), the number of increased parameters is 357.
After folding node (op_name: Range_356, op_type: Range), the number of increased parameters is 357.
After folding node (op_name: Mul_300, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: Mul_308, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: Mul_316, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: Mul_324, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: Mul_377, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: Mul_385, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: Mul_393, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: Mul_401, op_type: Mul), the number of increased parameters is 3.
After folding node (op_name: ConstantOfShape_298, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: ConstantOfShape_306, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: ConstantOfShape_314, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: ConstantOfShape_322, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: ConstantOfShape_375, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: ConstantOfShape_383, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: ConstantOfShape_391, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: ConstantOfShape_399, op_type: ConstantOfShape), the number of increased parameters is 2.
After folding node (op_name: Concat_166, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Concat_188, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Concat_201, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Concat_217, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Concat_226, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_240, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_241, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_242, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Concat_243, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Cast_269, op_type: Cast), the number of increased parameters is 0.
After folding node (op_name: Cast_276, op_type: Cast), the number of increased parameters is 0.
After folding node (op_name: Cast_283, op_type: Cast), the number of increased parameters is 0.
After folding node (op_name: Add_293, op_type: Add), the number of increased parameters is 0.
After folding node (op_name: Cast_346, op_type: Cast), the number of increased parameters is 0.
After folding node (op_name: Cast_353, op_type: Cast), the number of increased parameters is 0.
After folding node (op_name: Cast_360, op_type: Cast), the number of increased parameters is 0.
After folding node (op_name: Add_370, op_type: Add), the number of increased parameters is 0.
After folding node (op_name: Equal_301, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_304, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Equal_309, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_312, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Equal_317, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_320, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Equal_325, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_328, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Concat_329, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Concat_335, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Equal_378, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_381, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Equal_386, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_389, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Equal_394, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_397, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Equal_402, op_type: Equal), the number of increased parameters is 0.
After folding node (op_name: Unsqueeze_405, op_type: Unsqueeze), the number of increased parameters is 0.
After folding node (op_name: Concat_406, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Concat_412, op_type: Concat), the number of increased parameters is 0.
After folding node (op_name: Range_272, op_type: Range), the number of increased parameters is -2.
After folding node (op_name: Reshape_291, op_type: Reshape), the number of increased parameters is -2.
After folding node (op_name: Range_349, op_type: Range), the number of increased parameters is -2.
After folding node (op_name: Reshape_368, op_type: Reshape), the number of increased parameters is -2.
After folding node (op_name: Gather_233, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Gather_236, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Gather_239, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Gather_268, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Gather_275, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Gather_282, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Reshape_288, op_type: Reshape), the number of increased parameters is -4.
After folding node (op_name: Gather_345, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Gather_352, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Gather_359, op_type: Gather), the number of increased parameters is -4.
After folding node (op_name: Reshape_365, op_type: Reshape), the number of increased parameters is -4.
After folding node (op_name: Where_302, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Where_310, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Where_318, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Where_326, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Where_379, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Where_387, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Where_395, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Where_403, op_type: Where), the number of increased parameters is -4.
After folding node (op_name: Slice_165, op_type: Slice), the number of increased parameters is -5.
After folding node (op_name: Slice_187, op_type: Slice), the number of increased parameters is -5.
After folding node (op_name: Slice_200, op_type: Slice), the number of increased parameters is -5.
After folding node (op_name: Slice_216, op_type: Slice), the number of increased parameters is -5.
After folding node (op_name: Slice_225, op_type: Slice), the number of increased parameters is -5.
After folding node (op_name: Slice_334, op_type: Slice), the number of increased parameters is -7.
After folding node (op_name: Slice_411, op_type: Slice), the number of increased parameters is -7.
2026-04-20 11:01:29,269 file: model_builder.py func: model_builder line No: 38 End to optimize the onnx model.
2026-04-20 11:01:29,345 file: model_builder.py func: model_builder line No: 265 Saving model to: lraspp_99_torch110_optimized_float_model.onnx.
2026-04-20 11:01:29,345 file: model_builder.py func: model_builder line No: 35 Start to calibrate the model.
2026-04-20 11:01:29,630 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:Where_263_cast, type:Cast) does not support data type: int16
2026-04-20 11:01:29,658 file: calibration_data_set.py func: calibration_data_set line No: 111 input name: input, number_of_samples: 100
2026-04-20 11:01:29,658 file: calibration_data_set.py func: calibration_data_set line No: 123 There are 100 samples in the data set.
2026-04-20 11:01:29,659 file: infer_thresholds.py func: infer_thresholds line No: 84 Run calibration model with modelwise search method.
2026-04-20 11:01:30,323 file: base.py func: base line No: 138 Calibration using batch 8
2026-04-20 11:01:30,374 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:Where_263_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,612 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_263_cast, type:Cast) does not support data type: int16
2026-04-20 11:01:30,613 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_263_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,613 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:Where_263_equal_output_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,613 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_263_equal_output_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,613 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:ScatterND_337, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,613 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:ScatterND_337, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,614 file: tool_utils.py func: tool_utils line No: 321 The input1 of Node(name:ScatterND_337, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,614 file: tool_utils.py func: tool_utils line No: 321 The input1 of Node(name:ScatterND_337, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,614 file: tool_utils.py func: tool_utils line No: 321 The input2 of Node(name:ScatterND_337, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,614 file: tool_utils.py func: tool_utils line No: 321 The input2 of Node(name:ScatterND_337, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,614 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:ScatterND_337, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,615 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:ScatterND_337, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,615 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:Where_340_cast, type:Cast) does not support data type: int16
2026-04-20 11:01:30,615 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:Where_340_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,615 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_340_cast, type:Cast) does not support data type: int16
2026-04-20 11:01:30,615 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_340_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,615 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_340_equal, type:Equal) does not support data type: int16
2026-04-20 11:01:30,616 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:Where_340_equal_output_cast, type:Cast) does not support data type: int16
2026-04-20 11:01:30,616 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:Where_340_equal_output_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,616 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_340_equal_output_cast, type:Cast) does not support data type: int16
2026-04-20 11:01:30,616 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Where_340_equal_output_cast, type:Cast) does not support data type: int8
2026-04-20 11:01:30,616 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:ScatterND_414, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,616 file: tool_utils.py func: tool_utils line No: 321 The input0 of Node(name:ScatterND_414, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,616 file: tool_utils.py func: tool_utils line No: 321 The input1 of Node(name:ScatterND_414, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,617 file: tool_utils.py func: tool_utils line No: 321 The input1 of Node(name:ScatterND_414, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,617 file: tool_utils.py func: tool_utils line No: 321 The input2 of Node(name:ScatterND_414, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,617 file: tool_utils.py func: tool_utils line No: 321 The input2 of Node(name:ScatterND_414, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,617 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:ScatterND_414, type:ScatterND) does not support data type: int16
2026-04-20 11:01:30,617 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:ScatterND_414, type:ScatterND) does not support data type: int8
2026-04-20 11:01:30,617 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Resize_220, type:Resize) does not support data type: int16
2026-04-20 11:01:30,617 file: tool_utils.py func: tool_utils line No: 321 The output of Node(name:Resize_220) is int8, then requantized to int16
2026-04-20 11:01:40,827 file: ort.py func: ort line No: 207 Reset batch_size=1 and execute forward again...
2026-04-20 11:14:31,157 file: base.py func: base line No: 138 Calibration using batch 8
2026-04-20 11:14:33,617 file: ort.py func: ort line No: 207 Reset batch_size=1 and execute forward again...
2026-04-20 11:15:15,521 file: base.py func: base line No: 138 Calibration using batch 8
2026-04-20 11:15:26,593 file: ort.py func: ort line No: 207 Reset batch_size=1 and execute forward again...
2026-04-20 11:27:47,996 file: modelwise_search.py func: modelwise_search line No: 75 Select kl:num_bin=1024,max_num_bin=16384,per_channel method.
2026-04-20 11:27:47,996 file: modelwise_search.py func: modelwise_search line No: 79 Perchannel quantization is enabled.
2026-04-20 11:27:50,199 file: refine_threshold.py func: refine_threshold line No: 39 find illegal threshold in Add_258708_expand_dims_HzCalibration, it will be set default.
2026-04-20 11:27:50,486 file: model_builder.py func: model_builder line No: 38 End to calibrate the model.
2026-04-20 11:27:50,801 file: model_builder.py func: model_builder line No: 265 Saving model to: lraspp_99_torch110_calibrated_model.onnx.
2026-04-20 11:27:50,801 file: model_builder.py func: model_builder line No: 35 Start to quantize the model.
2026-04-20 11:27:54,932 file: constant_folding.py func: constant_folding line No: 66 Summary info for constant_folding:
2026-04-20 11:27:54,932 file: constant_folding.py func: constant_folding line No: 67 After constant_folding, the number of nodes has changed from 171 to 170.
2026-04-20 11:27:54,932 file: constant_folding.py func: constant_folding line No: 71 After constant_folding, the number of parameters has changed from 22325114 to 22325114.
2026-04-20 11:27:54,932 file: constant_folding.py func: constant_folding line No: 76 Detailed info for constant_folding:
2026-04-20 11:27:54,932 file: constant_folding.py func: constant_folding line No: 88 After folding node (op_name: Where_263_mul_x_NCHW2NHWC_LayoutConvert_Input1, op_type: Transpose), the number of increased parameters is 0.
2026-04-20 11:27:55,169 file: model_builder.py func: model_builder line No: 38 End to quantize the model.
2026-04-20 11:27:55,402 file: model_builder.py func: model_builder line No: 265 Saving model to: lraspp_99_torch110_quantized_model.onnx.
2026-04-20 11:27:55,403 file: model_builder.py func: model_builder line No: 35 Start to compile the model with march bayes-e.
2026-04-20 11:27:56,719 file: hybrid_build.py func: hybrid_build line No: 111 Compile submodel: torch-jit-export_subgraph_0
2026-04-20 11:27:56,855 file: hbdk_cc.py func: hbdk_cc line No: 126 hbdk-cc parameters:['--O3', '--debug', '--core-num', '1', '--fast', '--input-layout', 'NHWC', '--output-layout', 'NHWC', '--input-source', 'ddr']
2026-04-20 11:27:56,856 file: hbdk_cc.py func: hbdk_cc line No: 127 hbdk-cc command used:hbdk-cc -f hbir -m /tmp/tmp3_txy0ir/torch-jit-export_subgraph_0.hbir -o /tmp/tmp3_txy0ir/torch-jit-export_subgraph_0.hbm --march bayes-e --progressbar --O3 --debug --core-num 1 --fast --input-layout NHWC --output-layout NHWC --input-source ddr
2026-04-20 11:34:16,587 file: tool_utils.py func: tool_utils line No: 326 consumed time 379.708
2026-04-20 11:34:16,756 file: tool_utils.py func: tool_utils line No: 326 FPS=45.72, latency = 21874.5 us, DDR = 108467584 bytes (see torch-jit-export_subgraph_0.html)
2026-04-20 11:34:16,782 file: hybrid_build.py func: hybrid_build line No: 111 Compile submodel: torch-jit-export_subgraph_1
2026-04-20 11:34:16,908 file: hbdk_cc.py func: hbdk_cc line No: 126 hbdk-cc parameters:['--O3', '--debug', '--core-num', '1', '--fast', '--input-layout', 'NHWC', '--output-layout', 'NHWC', '--input-source', 'ddr,ddr']
2026-04-20 11:34:16,909 file: hbdk_cc.py func: hbdk_cc line No: 127 hbdk-cc command used:hbdk-cc -f hbir -m /tmp/tmp3_txy0ir/torch-jit-export_subgraph_1.hbir -o /tmp/tmp3_txy0ir/torch-jit-export_subgraph_1.hbm --march bayes-e --progressbar --O3 --debug --core-num 1 --fast --input-layout NHWC --output-layout NHWC --input-source ddr,ddr
2026-04-20 11:37:45,925 file: tool_utils.py func: tool_utils line No: 326 consumed time 208.991
2026-04-20 11:37:45,978 file: tool_utils.py func: tool_utils line No: 326 FPS=309.21, latency = 3234.0 us, DDR = 7143552 bytes (see torch-jit-export_subgraph_1.html)
2026-04-20 11:37:45,981 file: hybrid_build.py func: hybrid_build line No: 111 Compile submodel: torch-jit-export_subgraph_2
2026-04-20 11:37:46,054 file: hbdk_cc.py func: hbdk_cc line No: 126 hbdk-cc parameters:['--O3', '--debug', '--core-num', '1', '--fast', '--input-layout', 'NHWC', '--output-layout', 'NHWC', '--input-source', 'ddr,ddr,ddr']
2026-04-20 11:37:46,054 file: hbdk_cc.py func: hbdk_cc line No: 127 hbdk-cc command used:hbdk-cc -f hbir -m /tmp/tmp3_txy0ir/torch-jit-export_subgraph_2.hbir -o /tmp/tmp3_txy0ir/torch-jit-export_subgraph_2.hbm --march bayes-e --progressbar --O3 --debug --core-num 1 --fast --input-layout NHWC --output-layout NHWC --input-source ddr,ddr,ddr
2026-04-20 11:38:44,611 file: tool_utils.py func: tool_utils line No: 326 consumed time 58.5425
2026-04-20 11:38:44,663 file: tool_utils.py func: tool_utils line No: 326 FPS=89.51, latency = 11172.3 us, DDR = 95157888 bytes (see torch-jit-export_subgraph_2.html)
2026-04-20 11:38:45,023 file: model_builder.py func: model_builder line No: 38 End to compile the model with march bayes-e.
2026-04-20 11:38:51,561 file: print_info_dict.py func: print_info_dict line No: 72 The main quantized node information:
=================================================================================================================
Node ON Subgraph Type Cosine Similarity Threshold DataType
-----------------------------------------------------------------------------------------------------------------
Conv_0 BPU id(0) HzSQuantizedConv 0.986553 2.2984 int8
Mul_2 BPU id(0) HzLut 0.982618 11.5204 int8
Conv_3 BPU id(0) HzSQuantizedConv 0.990406 6.5663 int8
Conv_5 BPU id(0) HzSQuantizedConv 0.975953 9.86751 int8
Conv_7 BPU id(0) HzSQuantizedConv 0.974304 15.3085 int8
Conv_9 BPU id(0) HzSQuantizedConv 0.959030 7.18763 int8
Conv_11 BPU id(0) HzSQuantizedConv 0.962719 3.65214 int8
Conv_12 BPU id(0) HzSQuantizedConv 0.954578 8.41652 int8
Conv_14 BPU id(0) HzSQuantizedConv 0.943416 7.7668 int8
Conv_16 BPU id(0) HzSQuantizedConv 0.936729 3.44986 int8
Conv_18 BPU id(0) HzSQuantizedConv 0.954839 13.0062 int8
Conv_20 BPU id(0) HzSQuantizedConv 0.944272 3.26714 int8
GlobalAveragePool_22 BPU id(0) HzSQuantizedGlobalAveragePool 0.993927 6.53307 int8
Conv_23 BPU id(0) HzSQuantizedConv 0.989884 1.28699 int8
Conv_25 BPU id(0) HzSQuantizedConv 0.991844 2.15651 int8
HardSigmoid_26 BPU id(0) HzLut 0.999760 3.0 int8
Mul_27 BPU id(0) HzSElementwiseMul 0.961439 6.53307 int8
Conv_28 BPU id(0) HzSQuantizedConv 0.955945 5.40161 int8
Conv_29 BPU id(0) HzSQuantizedConv 0.955329 12.2378 int8
Conv_31 BPU id(0) HzSQuantizedConv 0.952980 8.71397 int8
GlobalAveragePool_33 BPU id(0) HzSQuantizedGlobalAveragePool 0.997060 6.15624 int8
Conv_34 BPU id(0) HzSQuantizedConv 0.997949 1.08985 int8
Conv_36 BPU id(0) HzSQuantizedConv 0.998952 2.13627 int8
HardSigmoid_37 BPU id(0) HzLut 0.999938 3.0 int8
Mul_38 BPU id(0) HzSElementwiseMul 0.966128 6.15624 int8
Conv_39 BPU id(0) HzSQuantizedConv 0.958394 3.84563 int8
Conv_41 BPU id(0) HzSQuantizedConv 0.962199 14.6263 int8
Conv_43 BPU id(0) HzSQuantizedConv 0.956062 8.403 int8
GlobalAveragePool_45 BPU id(0) HzSQuantizedGlobalAveragePool 0.997809 7.91869 int8
Conv_46 BPU id(0) HzSQuantizedConv 0.998193 1.21957 int8
Conv_48 BPU id(0) HzSQuantizedConv 0.997435 2.13308 int8
HardSigmoid_49 BPU id(0) HzLut 0.999917 3.0 int8
Mul_50 BPU id(0) HzSElementwiseMul 0.964402 7.91869 int8
Conv_51 BPU id(0) HzSQuantizedConv 0.955085 4.12897 int8
Conv_53 BPU id(0) HzSQuantizedConv 0.961900 17.1155 int8
Mul_55 BPU id(0) HzLut 0.963160 7.54118 int8
Conv_56 BPU id(0) HzSQuantizedConv 0.963110 6.7844 int8
Mul_58 BPU id(0) HzLut 0.961440 7.92997 int8
Conv_59 BPU id(0) HzSQuantizedConv 0.965863 7.92997 int8
Conv_60 BPU id(0) HzSQuantizedConv 0.964805 7.97945 int8
Mul_62 BPU id(0) HzLut 0.964762 7.46734 int8
Conv_63 BPU id(0) HzSQuantizedConv 0.964398 4.97091 int8
Mul_65 BPU id(0) HzLut 0.961959 8.2789 int8
Conv_66 BPU id(0) HzSQuantizedConv 0.963424 12.2686 int8
Conv_68 BPU id(0) HzSQuantizedConv 0.967719 12.6338 int8
Mul_70 BPU id(0) HzLut 0.967525 7.89655 int8
Conv_71 BPU id(0) HzSQuantizedConv 0.967249 6.31167 int8
Mul_73 BPU id(0) HzLut 0.964578 9.28673 int8
Conv_74 BPU id(0) HzSQuantizedConv 0.963028 10.58 int8
Conv_76 BPU id(0) HzSQuantizedConv 0.967019 15.7999 int8
Mul_78 BPU id(0) HzLut 0.966378 7.42885 int8
Conv_79 BPU id(0) HzSQuantizedConv 0.962902 5.51167 int8
Mul_81 BPU id(0) HzLut 0.963757 8.57854 int8
Conv_82 BPU id(0) HzSQuantizedConv 0.962496 8.87346 int8
Conv_84 BPU id(0) HzSQuantizedConv 0.971099 17.831 int8
Mul_86 BPU id(0) HzLut 0.971189 7.96112 int8
Conv_87 BPU id(0) HzSQuantizedConv 0.976725 6.11321 int8
Mul_89 BPU id(0) HzLut 0.974693 8.07926 int8
GlobalAveragePool_90 BPU id(0) HzSQuantizedGlobalAveragePool 0.991565 7.0649 int8
Conv_91 BPU id(0) HzSQuantizedConv 0.993154 0.734618 int8
Conv_93 BPU id(0) HzSQuantizedConv 0.992650 1.31863 int8
HardSigmoid_94 BPU id(0) HzLut 0.999965 3.0 int8
Mul_95 BPU id(0) HzSElementwiseMul 0.974305 7.0649 int8
Conv_96 BPU id(0) HzSQuantizedConv 0.979867 2.72911 int8
Conv_97 BPU id(0) HzSQuantizedConv 0.977957 8.07667 int8
Mul_99 BPU id(0) HzLut 0.977158 7.25696 int8
Conv_100 BPU id(0) HzSQuantizedConv 0.977867 4.82849 int8
Mul_102 BPU id(0) HzLut 0.976410 7.5785 int8
GlobalAveragePool_103 BPU id(0) HzSQuantizedGlobalAveragePool 0.990754 5.29187 int8
Conv_104 BPU id(0) HzSQuantizedConv 0.990275 0.762569 int8
Conv_106 BPU id(0) HzSQuantizedConv 0.991036 1.22081 int8
HardSigmoid_107 BPU id(0) HzLut 0.999961 3.0 int8
Mul_108 BPU id(0) HzSElementwiseMul 0.976040 5.29187 int8
Conv_109 BPU id(0) HzSQuantizedConv 0.977550 2.74255 int8
Conv_111 BPU id(0) HzSQuantizedConv 0.981876 12.1219 int8
Mul_113 BPU id(0) HzLut 0.980931 7.41744 int8
Conv_114 BPU id(0) HzSQuantizedConv 0.984248 4.31776 int8
Mul_116 BPU id(0) HzLut 0.981323 7.01304 int8
GlobalAveragePool_117 BPU id(0) HzSQuantizedGlobalAveragePool 0.991503 5.53688 int8
Conv_118 BPU id(0) HzSQuantizedConv 0.991524 0.828109 int8
Conv_120 BPU id(0) HzSQuantizedConv 0.991458 1.37029 int8
HardSigmoid_121 BPU id(0) HzLut 0.999946 3.0 int8
Mul_122 BPU id(0) HzSElementwiseMul 0.980876 5.53688 int8
Conv_123 BPU id(0) HzSQuantizedConv 0.984918 2.10283 int8
Conv_124 BPU id(0) HzSQuantizedConv 0.981502 5.73947 int8
Mul_126 BPU id(0) HzLut 0.978127 5.48624 int8
Conv_127 BPU id(0) HzSQuantizedConv 0.979374 3.65032 int8
Mul_129 BPU id(0) HzLut 0.976635 6.23468 int8
GlobalAveragePool_130 BPU id(0) HzSQuantizedGlobalAveragePool 0.986216 3.55395 int8
Conv_131 BPU id(0) HzSQuantizedConv 0.987970 0.924026 int8
Conv_133 BPU id(0) HzSQuantizedConv 0.989800 1.68722 int8
HardSigmoid_134 BPU id(0) HzLut 0.999947 3.0 int8
Mul_135 BPU id(0) HzSElementwiseMul 0.976136 3.55395 int8
Conv_136 BPU id(0) HzSQuantizedConv 0.973035 2.15599 int8
Conv_138 BPU id(0) HzSQuantizedConv 0.976796 7.28022 int8
Mul_140 BPU id(0) HzLut 0.973830 5.3063 int8
Conv_141 BPU id(0) HzSQuantizedConv 0.974648 5.02232 int8
Mul_143 BPU id(0) HzLut 0.970819 6.50338 int8
GlobalAveragePool_144 BPU id(0) HzSQuantizedGlobalAveragePool 0.982516 4.87036 int8
Conv_145 BPU id(0) HzSQuantizedConv 0.982393 0.929957 int8
Conv_147 BPU id(0) HzSQuantizedConv 0.983388 1.45901 int8
HardSigmoid_148 BPU id(0) HzLut 0.999915 3.0 int8
Mul_149 BPU id(0) HzSElementwiseMul 0.970485 4.87036 int8
Conv_150 BPU id(0) HzSQuantizedConv 0.957496 2.01668 int8
Conv_152 BPU id(0) HzSQuantizedConv 0.972063 9.2505 int8
Mul_154 BPU id(0) HzLut 0.968816 5.78079 int8
Conv_155 BPU id(0) HzSQuantizedConv 0.974796 4.99076 int8
GlobalAveragePool_157 BPU id(0) HzSQuantizedGlobalAveragePool 0.985728 4.99076 int8
Conv_158 BPU id(0) HzSQuantizedConv 0.999670 1.31951 int8
Sigmoid_159 BPU id(0) HzLut 0.999970 5.49897 int8
Mul_160 BPU id(0) HzSElementwiseMul 0.979894 9.05157 int8
Resize_169 BPU id(0) HzQuantizedRoiResize 0.983667 4.1394 int8
Conv_170 BPU id(0) HzSQuantizedConv 0.947539 8.41652 int8
Conv_171 BPU id(0) HzSQuantizedConv 0.991001 4.1394 int8
Conv_173 BPU id(0) HzSQuantizedConv 0.959445 8.41652 int8
Conv_174 BPU id(0) HzSQuantizedConv 0.992755 4.1394 int8
Conv_176 BPU id(0) HzSQuantizedConv 0.968049 8.41652 int8
598_calibrated_Requantize BPU id(0) HzRequantize -- -- int8
Concat_178 BPU id(0) Concat 0.979130 4.91939 int8
Conv_179 BPU id(0) HzSQuantizedConv 0.987551 4.91939 int8
Conv_181 BPU id(0) HzSQuantizedConv 0.985538 4.76962 int8
Resize_191 BPU id(0) HzQuantizedRoiResize 0.986748 4.16733 int8
Conv_192 BPU id(0) HzSQuantizedConv 0.986265 4.16733 int8
Conv_194 BPU id(0) HzSQuantizedConv 0.982295 4.43785 int8
Resize_204 BPU id(0) HzQuantizedRoiResize 0.982469 3.70782 int8
Conv_205 BPU id(0) HzSQuantizedConv 0.989918 3.70782 int8
Conv_207 BPU id(0) HzSQuantizedConv 0.982151 3.46615 int8
Conv_209 BPU id(0) HzSQuantizedConv 0.976264 4.16097 int8
Conv_211 BPU id(0) HzSQuantizedConv 0.990917 4.28882 int8
601_calibrated_Requantize BPU id(0) HzRequantize -- -- int16
Resize_220 BPU id(0) HzQuantizedRoiResize 0.990003 21.1521 int8
Resize_229 BPU id(0) HzQuantizedRoiResize 0.990926 32.0381 int8
Sigmoid_230 BPU id(0) HzLut2Layer 0.999924 11.0903 int16
665_Requantize BPU id(0) HzRequantize -- -- int8
Slice_249 BPU id(0) Slice 0.993033 21.1521 int16
ReduceMax_250 BPU id(0) HzQuantizedReduceMax 0.989577 21.1521 int16
Slice_255 BPU id(0) Slice 0.983892 32.0381 int8
ReduceMax_256 BPU id(0) HzQuantizedReduceMax 0.991725 32.0381 int8
Add_258 BPU id(0) HzSElementwiseAdd 0.991727 32.0381 int8
Less_259 BPU id(0) HzQuantizedComparison 0.969624 27.1529 int16
711_Cast CPU -- Cast -- -- float
Where_263_cast CPU -- Cast 0.982837 -- bool
Where_263_mul_x BPU id(1) HzSElementwiseMul 0.982776 127.0 int8
Where_263_equal BPU id(1) HzQuantizedComparison 0.969624 127.0 int8
Where_263_mul_y BPU id(1) HzSElementwiseMul 0.980290 127.0 int16
Where_263_add BPU id(1) HzSElementwiseAdd 0.982844 9916.99 int8
ScatterND_337 CPU -- ScatterND 0.997499 -- float
Gather_339 BPU id(2) Gather 0.999230 1254.88 int8
Where_340_cast CPU -- Cast 0.982837 -- bool
Where_340_mul_x BPU id(2) HzSElementwiseMul 0.968856 32767.0 int16
Where_340_equal BPU id(2) HzQuantizedComparison 0.969624 32767.0 int16
Where_340_mul_y BPU id(2) HzSElementwiseMul 0.969471 127.0 int8
Where_340_add BPU id(2) HzSElementwiseAdd 0.969493 23.9788 int8
ScatterND_414 CPU -- ScatterND 0.972476 -- float
2026-04-20 11:38:51,562 file: print_info_dict.py func: print_info_dict line No: 72 The quantized model output:
===============================================================================
Output Cosine Similarity L1 Distance L2 Distance Chebyshev Distance
-------------------------------------------------------------------------------
segmentation 0.972476 3329.117630 2.905576 10007.533203
depth 0.999924 0.003654 0.000019 0.140367
2026-04-20 11:38:51,570 file: model_builder.py func: model_builder line No: 38 End to Horizon NN Model Convert.
2026-04-20 11:38:51,659 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 601 start convert to *.bin file....
2026-04-20 11:38:51,675 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4326 ONNX model output num : 2
2026-04-20 11:38:54,229 file: layout_util.py func: layout_util line No: 15 set_featuremap_layout start
2026-04-20 11:38:54,229 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4060 model_deps_info: {'hb_mapper_version': '1.24.3', 'hbdk_version': '3.49.15', 'hbdk_runtime_version': ' 3.15.55.0', 'horizon_nn_version': '1.1.0', 'onnx_model': '/open_explorer/test_lraspp_model_0420/lraspp_99_torch110.onnx', 'march': 'bayes-e', 'layer_out_dump': False, 'log_level': 'DEBUG', 'working_dir': '/open_explorer/test_lraspp_model_0420/hb_mapper_output/lraspp_99_torch110', 'model_prefix': 'lraspp_99_torch110', 'node_info': {'711_Cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_263_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_263_equal': {'ON': 'BPU', 'InputType': 'int8', 'OutputType': 'int8'}, 'Where_263_equal_output_cast': {'ON': 'BPU', 'InputType': 'int8', 'OutputType': 'int8'}, 'ScatterND_337': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_equal': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_equal_output_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'ScatterND_414': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}}, 'input_names': ['input'], 'input_type_rt': ['featuremap'], 'input_space_and_range': ['regular'], 'input_type_train': ['featuremap'], 'input_layout_rt': ['NCHW'], 'input_layout_train': ['NCHW'], 'norm_type': ['no_preprocess'], 'scale_value': [''], 'mean_value': [''], 'input_shape': ['1x5x360x640'], 'input_batch': ['1'], 'cal_dir': ['/open_explorer/test_lraspp_model_0420/hb_calibration_data_f32'], 'cal_data_type': ['float32'], 'preprocess_on': False, 'calibration_type': 'default', 'per_channel': 'False', 'hbdk_params': {'hbdk_pass_through_params': '--O3 --debug --core-num 1 --fast ', 'input-source': {'input': 'ddr', '_default_value': 'ddr'}}, 'debug': True, 'compile_mode': 'latency'}
2026-04-20 11:38:54,229 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4183 ############# model deps info #############
2026-04-20 11:38:54,229 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4184 hb_mapper version : 1.24.3
2026-04-20 11:38:54,229 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4187 hbdk version : 3.49.15
2026-04-20 11:38:54,229 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4189 hbdk runtime version: 3.15.55.0
2026-04-20 11:38:54,229 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4192 horizon_nn version : 1.1.0
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4196 ############# model_parameters info #############
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4202 onnx_model : /open_explorer/test_lraspp_model_0420/lraspp_99_torch110.onnx
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4203 BPU march : bayes-e
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4204 layer_out_dump : False
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4205 log_level : DEBUG
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4206 working dir : /open_explorer/test_lraspp_model_0420/hb_mapper_output/lraspp_99_torch110
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4207 output_model_file_prefix: lraspp_99_torch110
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4226 node info : {'711_Cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_263_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_263_equal': {'ON': 'BPU', 'InputType': 'int8', 'OutputType': 'int8'}, 'Where_263_equal_output_cast': {'ON': 'BPU', 'InputType': 'int8', 'OutputType': 'int8'}, 'ScatterND_337': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_equal': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'Where_340_equal_output_cast': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}, 'ScatterND_414': {'ON': 'BPU', 'InputType': 'int16', 'OutputType': 'int16'}}
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4228 ############# input_parameters info #############
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4246 ------------------------------------------
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4248 ---------input info : input ---------
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4249 input_name : input
2026-04-20 11:38:54,230 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4250 input_type_rt : featuremap
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4252 input_space&range : regular
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4254 input_layout_rt : NCHW
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4255 input_type_train : featuremap
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4256 input_layout_train : NCHW
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4257 norm_type : no_preprocess
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4258 input_shape : 1x5x360x640
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4260 input_batch : 1
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4268 cal_data_dir : /open_explorer/test_lraspp_model_0420/hb_calibration_data_f32
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4270 cal_data_type : float32
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4271 ---------input info : input end -------
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4272 ------------------------------------------
2026-04-20 11:38:54,231 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4274 ############# calibration_parameters info #############
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4275 preprocess_on : False
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4276 calibration_type: : default
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4284 per_channel : False
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4303 ############# compiler_parameters info #############
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4305 debug : True
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4307 compile_mode : latency
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4310 hbdk_pass_through_params: --O3 --debug --core-num 1 --fast
2026-04-20 11:38:54,232 file: onnx2horizonrt.py func: onnx2horizonrt line No: 4310 input-source : {'input': 'ddr', '_default_value': 'ddr'}
2026-04-20 11:38:54,657 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 783 Convert to runtime bin file successfully!
2026-04-20 11:38:54,658 file: hb_mapper_makertbin.py func: hb_mapper_makertbin line No: 784 End Model Convert
根据排查手册，目前已做的排查：1. 对比各个中间模型的原始输出矩阵，发现在*_calibrated_model.onnx已产生较大的输出矩阵偏差，希望能提供一下帮助。
校准数据校验清单
文件格式检查
数据分布快速诊断
关键参数核对表
一致性验证（如果有训练时保存的 featuremap）
如果校验都通过，但精度仍低
后续反馈建议
关键问题定位
预期效果