This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

SK-TDA4VM: The questions about quantify onnx model on using TIDL rep

Part Number: SK-TDA4VM


 When i use the rep(https://github.com/TexasInstruments/edgeai-tidl-tools/tree/master) to quantify myself onnx model, some errors occur like above picture. Please help explain how to solve this? Thanks

  • Hi,

    Please share following details

    1. Which sdk version tag you are using ?

    2. Can you share model import logs with debug_trace set to 2 ?

  • Hello,

    the first: sdk version 09_01_06_00

    the second: 

    Available execution providers : ['TIDLExecutionProvider', 'TIDLCompilationProvider', 'CPUExecutionProvider']
    ['/home/root/TDA_calbration_frames_test/039273_india_2020-10-02T11-55-13.333126Z.jpg', '/home/root/TDA_calbration_frames_test/051549_india_2021-04-14T12-33-29.560214Z.jpg', '/home/root/TDA_calbration_frames_test/051609_india_2021-04-15T17-00-44.218167Z.jpg', '/home/root/TDA_calbration_frames_test/022224_india_2020-09-20T18-20-29.668663Z.jpg', '/home/root/TDA_calbration_frames_test/035502_india_2020-11-23T18-46-46.328718Z.jpg']

    Running 1 Models - ['clrnet']

    tidl_tools_path = /home/root/tidl_tools
    artifacts_folder = ../../../model-artifacts//clrnet/
    tidl_tensor_bits = 8
    debug_level = 2
    num_tidl_subgraphs = 16
    tidl_denylist =
    tidl_denylist_layer_name =
    tidl_denylist_layer_type =
    tidl_allowlist_layer_name =
    model_type =
    tidl_calibration_accuracy_level = 7
    tidl_calibration_options:num_frames_calibration = 5
    tidl_calibration_options:bias_calibration_iterations = 2
    mixed_precision_factor = -1.000000
    model_group_id = 0
    power_of_2_quantization = 2
    ONNX QDQ Enabled = 0
    enable_high_resolution_optimization = 0
    pre_batchnorm_fold = 1
    add_data_convert_ops = 3
    output_feature_16bit_names_list =
    m_params_16bit_names_list =
    reserved_compile_constraints_flag = 1601
    ti_internal_reserved_1 =


    ****** WARNING : Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options******

    Could not find const or initializer of layer Reshape_370 !!!

    Unsupported Onnx import data type : 0
    Could not find const or initializer of layer Reshape_370 !!!

    Unsupported Onnx import data type : 0
    Could not find const or initializer of layer Reshape_370 !!!

    Unsupported Onnx import data type : 0
    Supported TIDL layer type --- Conv -- Conv_0
    Supported TIDL layer type --- Relu -- Relu_1
    Supported TIDL layer type --- MaxPool -- MaxPool_2
    Supported TIDL layer type --- Conv -- Conv_3
    Supported TIDL layer type --- Relu -- Relu_4
    Supported TIDL layer type --- Conv -- Conv_5
    Supported TIDL layer type --- Add -- Add_6
    Supported TIDL layer type --- Relu -- Relu_7
    Supported TIDL layer type --- Conv -- Conv_8
    Supported TIDL layer type --- Relu -- Relu_9
    Supported TIDL layer type --- Conv -- Conv_10
    Supported TIDL layer type --- Add -- Add_11
    Supported TIDL layer type --- Relu -- Relu_12
    Supported TIDL layer type --- Conv -- Conv_16
    Supported TIDL layer type --- Conv -- Conv_13
    Supported TIDL layer type --- Relu -- Relu_14
    Supported TIDL layer type --- Conv -- Conv_15
    Supported TIDL layer type --- Add -- Add_17
    Supported TIDL layer type --- Relu -- Relu_18
    Supported TIDL layer type --- Conv -- Conv_19
    Supported TIDL layer type --- Relu -- Relu_20
    Supported TIDL layer type --- Conv -- Conv_21
    Supported TIDL layer type --- Add -- Add_22
    Supported TIDL layer type --- Relu -- Relu_23
    Supported TIDL layer type --- Conv -- Conv_27
    Supported TIDL layer type --- Conv -- Conv_24
    Supported TIDL layer type --- Relu -- Relu_25
    Supported TIDL layer type --- Conv -- Conv_26
    Supported TIDL layer type --- Add -- Add_28
    Supported TIDL layer type --- Relu -- Relu_29
    Supported TIDL layer type --- Conv -- Conv_30
    Supported TIDL layer type --- Relu -- Relu_31
    Supported TIDL layer type --- Conv -- Conv_32
    Supported TIDL layer type --- Add -- Add_33
    Supported TIDL layer type --- Relu -- Relu_34
    Supported TIDL layer type --- Conv -- Conv_38
    Supported TIDL layer type --- Conv -- Conv_35
    Supported TIDL layer type --- Relu -- Relu_36
    Supported TIDL layer type --- Conv -- Conv_37
    Supported TIDL layer type --- Add -- Add_39
    Supported TIDL layer type --- Relu -- Relu_40
    Supported TIDL layer type --- Conv -- Conv_41
    Supported TIDL layer type --- Relu -- Relu_42
    Supported TIDL layer type --- Conv -- Conv_43
    Supported TIDL layer type --- Add -- Add_44
    Supported TIDL layer type --- Relu -- Relu_45
    Supported TIDL layer type --- Conv -- Conv_48
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Slice -- Slice_62
    Supported TIDL layer type --- Conv -- Conv_47
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Gather -- Gather_54
    Unsupported (imports) TIDL layer type for ONNX op type --- Unsqueeze
    Supported TIDL layer type --- Gather -- Gather_51
    Unsupported (imports) TIDL layer type for ONNX op type --- Unsqueeze
    Supported TIDL layer type --- Concat -- Concat_57
    Supported TIDL layer type --- Concat -- Concat_64
    Supported TIDL layer type --- Resize -- Resize_67
    Supported TIDL layer type --- Add -- Add_68
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Slice -- Slice_82
    Supported TIDL layer type --- Conv -- Conv_46
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Gather -- Gather_74
    Unsupported (imports) TIDL layer type for ONNX op type --- Unsqueeze
    Supported TIDL layer type --- Gather -- Gather_71
    Unsupported (imports) TIDL layer type for ONNX op type --- Unsqueeze
    Supported TIDL layer type --- Concat -- Concat_77
    Supported TIDL layer type --- Concat -- Concat_84
    Supported TIDL layer type --- Resize -- Resize_87
    Supported TIDL layer type --- Add -- Add_88
    Supported TIDL layer type --- Conv -- Conv_89
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Gather -- Gather_94
    Unsupported (imports) TIDL layer type for ONNX op type --- Unsqueeze
    Supported TIDL layer type --- Concat -- Concat_96
    Unsupported (imports) TIDL layer type for ONNX op type --- Tile
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Slice -- Slice_553
    Supported TIDL layer type --- Gather -- Gather_500
    Unsupported (imports) TIDL layer type for ONNX op type --- Range
    Supported TIDL layer type --- Reshape -- Reshape_520
    Supported TIDL layer type --- Gather -- Gather_493
    Unsupported (imports) TIDL layer type for ONNX op type --- Range
    Supported TIDL layer type --- Reshape -- Reshape_518
    Supported TIDL layer type --- Add -- Add_521
    Supported TIDL layer type --- Gather -- Gather_507
    Unsupported (imports) TIDL layer type for ONNX op type --- Range
    Supported TIDL layer type --- Slice -- Slice_516
    Supported TIDL layer type --- Add -- Add_522
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Concat -- Concat_554
    Supported TIDL layer type --- Slice -- Slice_488
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Unsupported (imports) TIDL layer type for ONNX op type --- Tile
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Gather -- Gather_113
    Unsupported (imports) TIDL layer type for ONNX op type --- Unsqueeze
    Supported TIDL layer type --- Concat -- Concat_454
    Supported TIDL layer type --- Conv -- Conv_91
    Supported TIDL layer type --- Conv -- Conv_425
    Supported TIDL layer type --- Relu -- Relu_426
    Unsupported (imports) TIDL layer type for ONNX op type --- Shape
    Supported TIDL layer type --- Slice -- Slice_432
    Supported TIDL layer type --- Concat -- Concat_433
    Supported TIDL layer type --- Resiz

  • Hi 

    Can you provide the whole import log? There will be explanation about why these layers are not supported comes after what you pasted. 

    Regards,

    Adam