7c7533e4f2ae40f362748fc95d531ed3 /app/data/images/val/captured_image_10.jpg
2025-07-11 12:50:43,142 e[92mINFOe[0m log will be stored in /app/models/model_test/YOLO/hb_mapper_makertbin.log
2025-07-11 12:50:43,142 e[92mINFOe[0m Start hb_mapper…
2025-07-11 12:50:43,142 e[92mINFOe[0m hbdk version 3.49.15
2025-07-11 12:50:43,142 e[92mINFOe[0m horizon_nn version 1.1.0
2025-07-11 12:50:43,142 e[92mINFOe[0m hb_mapper version 1.24.3
2025-07-11 12:50:43,142 e[92mINFOe[0m Start Model Convert…
2025-07-11 12:50:43,145 e[92mINFOe[0m Using onnx model file: /app/logs/checker_output/.hb_check/quantized_model.onnx
2025-07-11 12:50:43,159 e[92mINFOe[0m Model has 1 inputs according to model file
2025-07-11 12:50:43,159 e[92mINFOe[0m Model name not given in yaml_file, using model name from model file: [‘images’]
2025-07-11 12:50:43,159 e[92mINFOe[0m Model input shape not given in yaml_file, using shape from model file: [[1, 3, 640, 640]]
2025-07-11 12:50:43,159 e[92mINFOe[0m nv12 input type rt received.
2025-07-11 12:50:43,160 e[93mWARNINGe[0m The calibration dir name suffix is not the same as the value float32 of the parameter cal_data_type, the parameter setting will prevail
2025-07-11 12:50:43,160 e[92mINFOe[0m custom_op does not exist, skipped
2025-07-11 12:50:43,160 e[93mWARNINGe[0m Input node images’s input_source not set, it will be set to pyramid by default
2025-07-11 12:50:43,160 e[93mWARNINGe[0m User input ‘preprocess_on’ will be deprecated. If you want to simplify the calibration process, please use ‘skip’ in ‘calibration_type’
2025-07-11 12:50:43,163 e[92mINFOe[0m *******************************************
2025-07-11 12:50:43,163 e[92mINFOe[0m First calibration picture name: captured_image_10.jpg
2025-07-11 12:50:43,163 e[92mINFOe[0m First calibration picture md5:
2025-07-11 12:50:43,183 e[92mINFOe[0m *******************************************
2025-07-11 12:50:46,845 e[92mINFOe[0m Start to Horizon NN Model Convert.
2025-07-11 12:50:46,859 e[92mINFOe[0m Loading horizon_nn debug methods:set()
2025-07-11 12:50:46,859 e[92mINFOe[0m The weight calibration parameters:
calibration_type: load
The activation calibration parameters:
calibration_type: load
The modelwise search parameters:
similarity: 0.995
metric: cosine-similarity
2025-07-11 12:50:46,859 e[92mINFOe[0m input images is from pyramid. Its layout is set to NHWC
2025-07-11 12:50:46,859 e[92mINFOe[0m The specified model compilation architecture: bernoulli2.
2025-07-11 12:50:46,859 e[92mINFOe[0m The specified model compilation optimization parameters: .
2025-07-11 12:50:46,859 e[92mINFOe[0m Start to prepare the onnx model.
2025-07-11 12:50:46,886 e[92mINFOe[0m Input ONNX Model Information:
ONNX IR version: 6
Opset version: [‘ai.onnx v11’, ‘horizon v1’]
Producer: horizon_nn v1.1.0
Domain: None
Version: None
Graph input:
images: shape=[1, 3, 640, 640], dtype=FLOAT32
Graph output:
small: shape=[1, 80, 80, 24], dtype=FLOAT32
medium: shape=[1, 40, 40, 24], dtype=FLOAT32
big: shape=[1, 20, 20, 24], dtype=FLOAT32
2025-07-11 12:50:47,155 e[92mINFOe[0m End to prepare the onnx model.
2025-07-11 12:50:47,181 e[92mINFOe[0m Saving model to: converted_model_original_float_model.onnx.
2025-07-11 12:50:47,182 e[92mINFOe[0m Start to optimize the onnx model.模型转化出现这种报错是什么问题?