Description
A clear and concise description of the bug or issue.
Environment
TensorRT Version: 8.2.2
GPU Type: rtx3090
Nvidia Driver Version:
CUDA Version: 11.2
CUDNN Version: 8.1.0
Operating System + Version: ubuntu 18.04
Python Version (if applicable): 3.6.13
TensorFlow Version (if applicable): 2.4
PyTorch Version (if applicable):
Baremetal or Container (if container which image + tag):
trtexec
when i use the trtexec tool to convert my onnx model to the trt model,it throw out a error “sequential/conv_lst_m2d/while/convolution_7: two inputs (data and weights) are allowed only in explicit-quantization mode.”
Here are the stdout and stderr
[03/21/2022-20:21:05] [I] === Model Options ===
[03/21/2022-20:21:05] [I] Format: ONNX
[03/21/2022-20:21:05] [I] Model: /home/liguangsheng/anomaly-detection/AND_trt/onnx2trt3/12_12_524_524_10.onnx
[03/21/2022-20:21:05] [I] Output:
[03/21/2022-20:21:05] [I] === Build Options ===
[03/21/2022-20:21:05] [I] Max batch: explicit batch
[03/21/2022-20:21:05] [I] Workspace: 16 MiB
[03/21/2022-20:21:05] [I] minTiming: 1
[03/21/2022-20:21:05] [I] avgTiming: 8
[03/21/2022-20:21:05] [I] Precision: FP32+INT8
[03/21/2022-20:21:05] [I] Calibration: Dynamic
[03/21/2022-20:21:05] [I] Refit: Disabled
[03/21/2022-20:21:05] [I] Sparsity: Disabled
[03/21/2022-20:21:05] [I] Safe mode: Disabled
[03/21/2022-20:21:05] [I] DirectIO mode: Disabled
[03/21/2022-20:21:05] [I] Restricted mode: Disabled
[03/21/2022-20:21:05] [I] Save engine: 1.trt
[03/21/2022-20:21:05] [I] Load engine:
[03/21/2022-20:21:05] [I] Profiling verbosity: 0
[03/21/2022-20:21:05] [I] Tactic sources: Using default tactic sources
[03/21/2022-20:21:05] [I] timingCacheMode: local
[03/21/2022-20:21:05] [I] timingCacheFile:
[03/21/2022-20:21:05] [I] Input(s)s format: fp32:CHW
[03/21/2022-20:21:05] [I] Output(s)s format: fp32:CHW
[03/21/2022-20:21:05] [I] Input build shapes: model
[03/21/2022-20:21:05] [I] Input calibration shapes: model
[03/21/2022-20:21:05] [I] === System Options ===
[03/21/2022-20:21:05] [I] Device: 0
[03/21/2022-20:21:05] [I] DLACore:
[03/21/2022-20:21:05] [I] Plugins:
[03/21/2022-20:21:05] [I] === Inference Options ===
[03/21/2022-20:21:05] [I] Batch: Explicit
[03/21/2022-20:21:05] [I] Input inference shapes: model
[03/21/2022-20:21:05] [I] Iterations: 10
[03/21/2022-20:21:05] [I] Duration: 3s (+ 200ms warm up)
[03/21/2022-20:21:05] [I] Sleep time: 0ms
[03/21/2022-20:21:05] [I] Idle time: 0ms
[03/21/2022-20:21:05] [I] Streams: 1
[03/21/2022-20:21:05] [I] ExposeDMA: Disabled
[03/21/2022-20:21:05] [I] Data transfers: Enabled
[03/21/2022-20:21:05] [I] Spin-wait: Disabled
[03/21/2022-20:21:05] [I] Multithreading: Disabled
[03/21/2022-20:21:05] [I] CUDA Graph: Disabled
[03/21/2022-20:21:05] [I] Separate profiling: Disabled
[03/21/2022-20:21:05] [I] Time Deserialize: Disabled
[03/21/2022-20:21:05] [I] Time Refit: Disabled
[03/21/2022-20:21:05] [I] Skip inference: Disabled
[03/21/2022-20:21:05] [I] Inputs:
[03/21/2022-20:21:05] [I] === Reporting Options ===
[03/21/2022-20:21:05] [I] Verbose: Enabled
[03/21/2022-20:21:05] [I] Averages: 10 inferences
[03/21/2022-20:21:05] [I] Percentile: 99
[03/21/2022-20:21:05] [I] Dump refittable layers:Disabled
[03/21/2022-20:21:05] [I] Dump output: Disabled
[03/21/2022-20:21:05] [I] Profile: Disabled
[03/21/2022-20:21:05] [I] Export timing to JSON file:
[03/21/2022-20:21:05] [I] Export output to JSON file:
[03/21/2022-20:21:05] [I] Export profile to JSON file:
[03/21/2022-20:21:05] [I]
[03/21/2022-20:21:05] [I] === Device Information ===
[03/21/2022-20:21:05] [I] Selected Device: NVIDIA GeForce RTX 3090
[03/21/2022-20:21:05] [I] Compute Capability: 8.6
[03/21/2022-20:21:05] [I] SMs: 82
[03/21/2022-20:21:05] [I] Compute Clock Rate: 1.695 GHz
[03/21/2022-20:21:05] [I] Device Global Memory: 24268 MiB
[03/21/2022-20:21:05] [I] Shared Memory per SM: 100 KiB
[03/21/2022-20:21:05] [I] Memory Bus Width: 384 bits (ECC disabled)
[03/21/2022-20:21:05] [I] Memory Clock Rate: 9.751 GHz
[03/21/2022-20:21:05] [I]
[03/21/2022-20:21:05] [I] TensorRT version: 8.2.2
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::GridAnchor_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::GridAnchorRect_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::NMS_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::Reorg_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::Region_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::Clip_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::LReLU_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::PriorBox_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::Normalize_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::ScatterND version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::RPROI_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::BatchedNMS_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::BatchedNMSDynamic_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::FlattenConcat_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::CropAndResize version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::DetectionLayer_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::EfficientNMS_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::EfficientNMS_ONNX_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::EfficientNMS_TFTRT_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::Proposal version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::ProposalLayer_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::PyramidROIAlign_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::ResizeNearest_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::Split version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::SpecialSlice_TRT version 1
[03/21/2022-20:21:05] [V] [TRT] Registered plugin creator - ::InstanceNormalization_TRT version 1
[03/21/2022-20:21:06] [I] [TRT] [MemUsageChange] Init CUDA: CPU +460, GPU +0, now: CPU 471, GPU 16992 (MiB)
[03/21/2022-20:21:06] [I] [TRT] [MemUsageSnapshot] Begin constructing builder kernel library: CPU 471 MiB, GPU 16992 MiB[03/21/2022-20:21:07] [I] [TRT] [MemUsageSnapshot] End constructing builder kernel library: CPU 625 MiB, GPU 17036 MiB
[03/21/2022-20:21:07] [I] Start parsing network model
[03/21/2022-20:21:07] [I] [TRT] ----------------------------------------------------------------
[03/21/2022-20:21:07] [I] [TRT] Input filename: /home/liguangsheng/anomaly-detection/AND_trt/onnx2trt3/12_12_524_524_10.onnx
[03/21/2022-20:21:07] [I] [TRT] ONNX IR version: 0.0.5
[03/21/2022-20:21:07] [I] [TRT] Opset version: 10
[03/21/2022-20:21:07] [I] [TRT] Producer name: tf2onnx
[03/21/2022-20:21:07] [I] [TRT] Producer version: 1.9.3
[03/21/2022-20:21:07] [I] [TRT] Domain:
[03/21/2022-20:21:07] [I] [TRT] Model version: 0
[03/21/2022-20:21:07] [I] [TRT] Doc string:
[03/21/2022-20:21:07] [I] [TRT] ----------------------------------------------------------------
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::GridAnchor_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::GridAnchorRect_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::NMS_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::Reorg_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::Region_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::Clip_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::LReLU_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::PriorBox_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::Normalize_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::ScatterND version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::RPROI_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::BatchedNMS_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::BatchedNMSDynamic_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::FlattenConcat_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::CropAndResize version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::DetectionLayer_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::EfficientNMS_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::EfficientNMS_ONNX_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::EfficientNMS_TFTRT_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::Proposal version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::ProposalLayer_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::PyramidROIAlign_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::ResizeNearest_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::Split version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::SpecialSlice_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Plugin creator already registered - ::InstanceNormalization_TRT version 1
[03/21/2022-20:21:07] [V] [TRT] Adding network input: conv3d_input:0 with dtype: float32, dimensions: (-1, 256, 256, 16, 1)
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: conv3d_input:0 for ONNX tensor: conv3d_input:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: zero__243
[03/21/2022-20:21:07] [W] [TRT] onnx2trt_utils.cpp:366: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_2/zeros:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_2/split_2/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_2/split_1/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_2/split/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_1/zeros:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_1/while/maximum_iterations:0
[03/21/2022-20:21:07] [V] [TRT] Weight at index 0: 9223372036854775807 is out of range. Clamping to: 2147483647
[03/21/2022-20:21:07] [W] [TRT] onnx2trt_utils.cpp:392: One or more weights outside the range of INT32 was clamped
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_1/split_2/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_1/split_1/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d_1/split/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/zeros:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/time:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/strided_slice__48
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/split_2/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/split_1/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/split/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d_transpose_1/mul_1/y:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d_transpose_1/conv3d_transpose_const_slice_axes__410
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d_transpose_1/conv3d_transpose/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d_transpose/mul/y:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d_transpose/conv3d_transpose/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d_1/Conv3D/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d_1/BiasAdd/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d/Conv3D/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv3d/BiasAdd/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: new_shape__536
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: indice__534
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_starts__374
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_starts__307
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__911
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__910
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__908
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__907
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__906
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__905
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__1149
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__1148
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_ends__353
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_ends__349
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_axes__350
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: Transpose__500_shape__1027
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv3d/Conv3D__233 [Reshape]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: conv3d_input:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: new_shape__536
[03/21/2022-20:21:07] [V] [TRT] sequential/conv3d/Conv3D__233 [Reshape] inputs: [conv3d_input:0 -> (-1, 256, 256, 16, 1)[FLOAT]], [new_shape__536 -> (5)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv3d/Conv3D__233 for ONNX node: sequential/conv3d/Conv3D__233
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv3d/Conv3D__233:0 for ONNX tensor: sequential/conv3d/Conv3D__233:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv3d/Conv3D__233 [Reshape] outputs: [sequential/conv3d/Conv3D__233:0 -> (-1, 1, 256, 256, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less [Less]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: const_fold_opt__1149
[03/21/2022-20:21:07] [V] [TRT] Searching for input: const_fold_opt__1148
[03/21/2022-20:21:07] [V] [TRT] sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less [Less] inputs: [const_fold_opt__1149 -> ()[FLOAT]], [const_fold_opt__1148 -> ()[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: const_fold_opt__1149 for ONNX node: const_fold_opt__1149
[03/21/2022-20:21:07] [V] [TRT] Registering layer: const_fold_opt__1148 for ONNX node: const_fold_opt__1148
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less for ONNX node: sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less:0 for ONNX tensor: sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less:0
[03/21/2022-20:21:07] [V] [TRT] sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less [Less] outputs: [sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less:0 -> ()[BOOL]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: Conv__437 [Conv]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d/Conv3D__233:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d/Conv3D/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d/BiasAdd/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Conv__437 [Conv] inputs: [sequential/conv3d/Conv3D__233:0 -> (-1, 1, 256, 256, 16)[FLOAT]], [sequential/conv3d/Conv3D/ReadVariableOp:0 -> (128, 1, 11, 11, 1)[FLOAT]], [sequential/conv3d/BiasAdd/ReadVariableOp:0 -> (128)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Convolution input dimensions: (-1, 1, 256, 256, 16)
[03/21/2022-20:21:07] [V] [TRT] Registering layer: Conv__437 for ONNX node: Conv__437
[03/21/2022-20:21:07] [V] [TRT] Using kernel: (11, 11, 1), strides: (4, 4, 1), prepadding: (3, 3, 0), postpadding: (4, 4, 0), dilations: (1, 1, 1), numOutputs: 128
[03/21/2022-20:21:07] [V] [TRT] Convolution output dimensions: (-1, 128, 64, 64, 16)
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: Conv__437:0 for ONNX tensor: Conv__437:0
[03/21/2022-20:21:07] [V] [TRT] Conv__437 [Conv] outputs: [Conv__437:0 -> (-1, 128, 64, 64, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv3d/Tanh [Tanh]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: Conv__437:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv3d/Tanh [Tanh] inputs: [Conv__437:0 -> (-1, 128, 64, 64, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv3d/Tanh for ONNX node: sequential/conv3d/Tanh
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv3d/Tanh:0 for ONNX tensor: sequential/conv3d/Tanh:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv3d/Tanh [Tanh] outputs: [sequential/conv3d/Tanh:0 -> (-1, 128, 64, 64, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: Conv__439 [Conv]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d/Tanh:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d_1/Conv3D/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d_1/BiasAdd/ReadVariableOp:0
[03/21/2022-20:21:07] [V] [TRT] Conv__439 [Conv] inputs: [sequential/conv3d/Tanh:0 -> (-1, 128, 64, 64, 16)[FLOAT]], [sequential/conv3d_1/Conv3D/ReadVariableOp:0 -> (64, 128, 5, 5, 1)[FLOAT]], [sequential/conv3d_1/BiasAdd/ReadVariableOp:0 -> (64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Convolution input dimensions: (-1, 128, 64, 64, 16)
[03/21/2022-20:21:07] [V] [TRT] Registering layer: Conv__439 for ONNX node: Conv__439
[03/21/2022-20:21:07] [V] [TRT] Using kernel: (5, 5, 1), strides: (2, 2, 1), prepadding: (1, 1, 0), postpadding: (2, 2, 0), dilations: (1, 1, 1), numOutputs: 64
[03/21/2022-20:21:07] [V] [TRT] Convolution output dimensions: (-1, 64, 32, 32, 16)
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: Conv__439:0 for ONNX tensor: Conv__439:0
[03/21/2022-20:21:07] [V] [TRT] Conv__439 [Conv] outputs: [Conv__439:0 -> (-1, 64, 32, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv3d_1/Tanh [Tanh]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: Conv__439:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv3d_1/Tanh [Tanh] inputs: [Conv__439:0 -> (-1, 64, 32, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv3d_1/Tanh for ONNX node: sequential/conv3d_1/Tanh
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv3d_1/Tanh:0 for ONNX tensor: sequential/conv3d_1/Tanh:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv3d_1/Tanh [Tanh] outputs: [sequential/conv3d_1/Tanh:0 -> (-1, 64, 32, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/transpose [Transpose]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d_1/Tanh:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/transpose [Transpose] inputs: [sequential/conv3d_1/Tanh:0 -> (-1, 64, 32, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/transpose for ONNX node: sequential/conv_lst_m2d/transpose
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/transpose:0 for ONNX tensor: sequential/conv_lst_m2d/transpose:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/transpose [Transpose] outputs: [sequential/conv_lst_m2d/transpose:0 -> (32, -1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: Cast__241 [Cast]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv3d_1/Tanh:0
[03/21/2022-20:21:07] [V] [TRT] Cast__241 [Cast] inputs: [sequential/conv3d_1/Tanh:0 -> (-1, 64, 32, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Casting to type: int32
[03/21/2022-20:21:07] [V] [TRT] Registering layer: Cast__241 for ONNX node: Cast__241
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: Cast__241:0 for ONNX tensor: Cast__241:0
[03/21/2022-20:21:07] [V] [TRT] Cast__241 [Cast] outputs: [Cast__241:0 -> (-1, 64, 32, 32, 16)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: Mul__244 [Mul]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: Cast__241:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: zero__243
[03/21/2022-20:21:07] [V] [TRT] Mul__244 [Mul] inputs: [Cast__241:0 -> (-1, 64, 32, 32, 16)[INT32]], [zero__243 -> ()[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: zero__243 for ONNX node: zero__243
[03/21/2022-20:21:07] [V] [TRT] Registering layer: Mul__244 for ONNX node: Mul__244
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: Mul__244:0 for ONNX tensor: Mul__244:0
[03/21/2022-20:21:07] [V] [TRT] Mul__244 [Mul] outputs: [Mul__244:0 -> (-1, 64, 32, 32, 16)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/zeros_like [Cast]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: Mul__244:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/zeros_like [Cast] inputs: [Mul__244:0 -> (-1, 64, 32, 32, 16)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Casting to type: float32
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/zeros_like for ONNX node: sequential/conv_lst_m2d/zeros_like
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/zeros_like:0 for ONNX tensor: sequential/conv_lst_m2d/zeros_like:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/zeros_like [Cast] outputs: [sequential/conv_lst_m2d/zeros_like:0 -> (-1, 64, 32, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/Sum [ReduceSum]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/zeros_like:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/Sum [ReduceSum] inputs: [sequential/conv_lst_m2d/zeros_like:0 -> (-1, 64, 32, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/Sum for ONNX node: sequential/conv_lst_m2d/Sum
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/Sum:0 for ONNX tensor: sequential/conv_lst_m2d/Sum:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/Sum [ReduceSum] outputs: [sequential/conv_lst_m2d/Sum:0 -> (-1, 64, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/convolution [Conv]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/Sum:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/zeros:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/convolution [Conv] inputs: [sequential/conv_lst_m2d/Sum:0 -> (-1, 64, 32, 16)[FLOAT]], [sequential/conv_lst_m2d/zeros:0 -> (64, 64, 3, 3)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Convolution input dimensions: (-1, 64, 32, 16)
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/convolution for ONNX node: sequential/conv_lst_m2d/convolution
[03/21/2022-20:21:07] [V] [TRT] Using kernel: (3, 3), strides: (1, 1), prepadding: (1, 1), postpadding: (1, 1), dilations: (1, 1), numOutputs: 64
[03/21/2022-20:21:07] [V] [TRT] Convolution output dimensions: (-1, 64, 32, 16)
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/convolution:0 for ONNX tensor: sequential/conv_lst_m2d/convolution:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/convolution [Conv] outputs: [sequential/conv_lst_m2d/convolution:0 -> (-1, 64, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: Transpose__445 [Transpose]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/convolution:0
[03/21/2022-20:21:07] [V] [TRT] Transpose__445 [Transpose] inputs: [sequential/conv_lst_m2d/convolution:0 -> (-1, 64, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: Transpose__445 for ONNX node: Transpose__445
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: Transpose__445:0 for ONNX tensor: Transpose__445:0
[03/21/2022-20:21:07] [V] [TRT] Transpose__445 [Transpose] outputs: [Transpose__445:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while_loop [Loop]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d_1/while/maximum_iterations:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/time:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: Transpose__445:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: Transpose__445:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/strided_slice__48
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/split/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/split_1/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/split_2/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while_loop [Loop] inputs: [sequential/conv_lst_m2d_1/while/maximum_iterations:0 -> ()[INT32]], [sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less:0 -> ()[BOOL]], [sequential/conv_lst_m2d/time:0 -> ()[INT32]], [Transpose__445:0 -> (-1, 32, 16, 64)[FLOAT]], [Transpose__445:0 -> (-1, 32, 16, 64)[FLOAT]], [sequential/conv_lst_m2d/strided_slice__48 -> ()[INT32]], [sequential/conv_lst_m2d/split/ReadVariableOp/resource:0 -> (3, 3, 64, 256)[FLOAT]], [sequential/conv_lst_m2d/split_1/ReadVariableOp/resource:0 -> (3, 3, 64, 256)[FLOAT]], [sequential/conv_lst_m2d/split_2/ReadVariableOp/resource:0 -> (256)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d_1/while/maximum_iterations:0 for ONNX node: sequential/conv_lst_m2d_1/while/maximum_iterations:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_sequential_conv_lst_m2d_while_loop_counter:0 for ONNX tensor: sequential_conv_lst_m2d_while_sequential_conv_lst_m2d_while_loop_counter:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: cond__250:0 for ONNX tensor: cond__250:0
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/time:0 for ONNX node: sequential/conv_lst_m2d/time:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_placeholder:0 for ONNX tensor: sequential_conv_lst_m2d_while_placeholder:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_placeholder_2:0 for ONNX tensor: sequential_conv_lst_m2d_while_placeholder_2:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_placeholder_3:0 for ONNX tensor: sequential_conv_lst_m2d_while_placeholder_3:0
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/strided_slice__48 for ONNX node: sequential/conv_lst_m2d/strided_slice__48
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_sequential_conv_lst_m2d_strided_slice_0:0 for ONNX tensor: sequential_conv_lst_m2d_while_sequential_conv_lst_m2d_strided_slice_0:0
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/split/ReadVariableOp/resource:0 for ONNX node: sequential/conv_lst_m2d/split/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_split_readvariableop_resource_0:0 for ONNX tensor: sequential_conv_lst_m2d_while_split_readvariableop_resource_0:0
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/split_1/ReadVariableOp/resource:0 for ONNX node: sequential/conv_lst_m2d/split_1/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_split_1_readvariableop_resource_0:0 for ONNX tensor: sequential_conv_lst_m2d_while_split_1_readvariableop_resource_0:0
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/split_2/ReadVariableOp/resource:0 for ONNX node: sequential/conv_lst_m2d/split_2/ReadVariableOp/resource:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential_conv_lst_m2d_while_split_2_readvariableop_resource_0:0 for ONNX tensor: sequential_conv_lst_m2d_while_split_2_readvariableop_resource_0:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/while/add_8/y:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/while/Const_7:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/while/Const_5:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: sequential/conv_lst_m2d/while/Const_3:0
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__947
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__946
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__945
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__943
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__942
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__941
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__937
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__934
[03/21/2022-20:21:07] [V] [TRT] Importing initializer: const_fold_opt__933
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/split_2 [Split]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_split_2_readvariableop_resource_0:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/split_2 [Split] inputs: [sequential_conv_lst_m2d_while_split_2_readvariableop_resource_0:0 -> (256)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_2 for ONNX node: sequential/conv_lst_m2d/while/split_2
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_2_0 for ONNX node: sequential/conv_lst_m2d/while/split_2
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_2_1 for ONNX node: sequential/conv_lst_m2d/while/split_2
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_2_2 for ONNX node: sequential/conv_lst_m2d/while/split_2
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_2:0 for ONNX tensor: sequential/conv_lst_m2d/while/split_2:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_2:1 for ONNX tensor: sequential/conv_lst_m2d/while/split_2:1
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_2:2 for ONNX tensor: sequential/conv_lst_m2d/while/split_2:2
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_2:3 for ONNX tensor: sequential/conv_lst_m2d/while/split_2:3
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/split_2 [Split] outputs: [sequential/conv_lst_m2d/while/split_2:0 -> (64)[FLOAT]], [sequential/conv_lst_m2d/while/split_2:1 -> (64)[FLOAT]], [sequential/conv_lst_m2d/while/split_2:2 -> (64)[FLOAT]], [sequential/conv_lst_m2d/while/split_2:3 -> (64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/split_1 [Split]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_split_1_readvariableop_resource_0:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/split_1 [Split] inputs: [sequential_conv_lst_m2d_while_split_1_readvariableop_resource_0:0 -> (3, 3, 64, 256)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_1 for ONNX node: sequential/conv_lst_m2d/while/split_1
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_1_3 for ONNX node: sequential/conv_lst_m2d/while/split_1
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_1_4 for ONNX node: sequential/conv_lst_m2d/while/split_1
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_1_5 for ONNX node: sequential/conv_lst_m2d/while/split_1
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_1:0 for ONNX tensor: sequential/conv_lst_m2d/while/split_1:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_1:1 for ONNX tensor: sequential/conv_lst_m2d/while/split_1:1
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_1:2 for ONNX tensor: sequential/conv_lst_m2d/while/split_1:2
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split_1:3 for ONNX tensor: sequential/conv_lst_m2d/while/split_1:3
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/split_1 [Split] outputs: [sequential/conv_lst_m2d/while/split_1:0 -> (3, 3, 64, 64)[FLOAT]], [sequential/conv_lst_m2d/while/split_1:1 -> (3, 3, 64, 64)[FLOAT]], [sequential/conv_lst_m2d/while/split_1:2 -> (3, 3, 64, 64)[FLOAT]], [sequential/conv_lst_m2d/while/split_1:3 -> (3, 3, 64, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/split [Split]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_split_readvariableop_resource_0:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/split [Split] inputs: [sequential_conv_lst_m2d_while_split_readvariableop_resource_0:0 -> (3, 3, 64, 256)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split for ONNX node: sequential/conv_lst_m2d/while/split
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_6 for ONNX node: sequential/conv_lst_m2d/while/split
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_7 for ONNX node: sequential/conv_lst_m2d/while/split
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/split_8 for ONNX node: sequential/conv_lst_m2d/while/split
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split:0 for ONNX tensor: sequential/conv_lst_m2d/while/split:0
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split:1 for ONNX tensor: sequential/conv_lst_m2d/while/split:1
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split:2 for ONNX tensor: sequential/conv_lst_m2d/while/split:2
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/split:3 for ONNX tensor: sequential/conv_lst_m2d/while/split:3
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/split [Split] outputs: [sequential/conv_lst_m2d/while/split:0 -> (3, 3, 64, 64)[FLOAT]], [sequential/conv_lst_m2d/while/split:1 -> (3, 3, 64, 64)[FLOAT]], [sequential/conv_lst_m2d/while/split:2 -> (3, 3, 64, 64)[FLOAT]], [sequential/conv_lst_m2d/while/split:3 -> (3, 3, 64, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/ones_like_1/Shape [Shape]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_placeholder_2:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1/Shape [Shape] inputs: [sequential_conv_lst_m2d_while_placeholder_2:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/ones_like_1/Shape for ONNX node: sequential/conv_lst_m2d/while/ones_like_1/Shape
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/ones_like_1/Shape:0 for ONNX tensor: sequential/conv_lst_m2d/while/ones_like_1/Shape:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1/Shape [Shape] outputs: [sequential/conv_lst_m2d/while/ones_like_1/Shape:0 -> (4)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/ones_like_1/Shape__113 [Cast]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/ones_like_1/Shape:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1/Shape__113 [Cast] inputs: [sequential/conv_lst_m2d/while/ones_like_1/Shape:0 -> (4)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Casting to type: int32
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/ones_like_1/Shape__113 for ONNX node: sequential/conv_lst_m2d/while/ones_like_1/Shape__113
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/ones_like_1/Shape__113:0 for ONNX tensor: sequential/conv_lst_m2d/while/ones_like_1/Shape__113:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1/Shape__113 [Cast] outputs: [sequential/conv_lst_m2d/while/ones_like_1/Shape__113:0 -> (4)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/ones_like_1__114 [Cast]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/ones_like_1/Shape__113:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1__114 [Cast] inputs: [sequential/conv_lst_m2d/while/ones_like_1/Shape__113:0 -> (4)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Casting to type: int32
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/ones_like_1__114 for ONNX node: sequential/conv_lst_m2d/while/ones_like_1__114
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/ones_like_1__114:0 for ONNX tensor: sequential/conv_lst_m2d/while/ones_like_1__114:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1__114 [Cast] outputs: [sequential/conv_lst_m2d/while/ones_like_1__114:0 -> (4)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/ones_like_1 [ConstantOfShape]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/ones_like_1__114:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1 [ConstantOfShape] inputs: [sequential/conv_lst_m2d/while/ones_like_1__114:0 -> (4)[INT32]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/ones_like_1 for ONNX node: sequential/conv_lst_m2d/while/ones_like_1
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/ones_like_1:0 for ONNX tensor: sequential/conv_lst_m2d/while/ones_like_1:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/ones_like_1 [ConstantOfShape] outputs: [sequential/conv_lst_m2d/while/ones_like_1:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/mul_7 [Mul]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_placeholder_2:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/ones_like_1:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_7 [Mul] inputs: [sequential_conv_lst_m2d_while_placeholder_2:0 -> (-1, 32, 16, 64)[FLOAT]], [sequential/conv_lst_m2d/while/ones_like_1:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/mul_7 for ONNX node: sequential/conv_lst_m2d/while/mul_7
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/mul_7:0 for ONNX tensor: sequential/conv_lst_m2d/while/mul_7:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_7 [Mul] outputs: [sequential/conv_lst_m2d/while/mul_7:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/mul_6 [Mul]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_placeholder_2:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/ones_like_1:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_6 [Mul] inputs: [sequential_conv_lst_m2d_while_placeholder_2:0 -> (-1, 32, 16, 64)[FLOAT]], [sequential/conv_lst_m2d/while/ones_like_1:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/mul_6 for ONNX node: sequential/conv_lst_m2d/while/mul_6
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/mul_6:0 for ONNX tensor: sequential/conv_lst_m2d/while/mul_6:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_6 [Mul] outputs: [sequential/conv_lst_m2d/while/mul_6:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/mul_5 [Mul]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_placeholder_2:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/ones_like_1:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_5 [Mul] inputs: [sequential_conv_lst_m2d_while_placeholder_2:0 -> (-1, 32, 16, 64)[FLOAT]], [sequential/conv_lst_m2d/while/ones_like_1:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/mul_5 for ONNX node: sequential/conv_lst_m2d/while/mul_5
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/mul_5:0 for ONNX tensor: sequential/conv_lst_m2d/while/mul_5:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_5 [Mul] outputs: [sequential/conv_lst_m2d/while/mul_5:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/mul_4 [Mul]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential_conv_lst_m2d_while_placeholder_2:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/ones_like_1:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_4 [Mul] inputs: [sequential_conv_lst_m2d_while_placeholder_2:0 -> (-1, 32, 16, 64)[FLOAT]], [sequential/conv_lst_m2d/while/ones_like_1:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/mul_4 for ONNX node: sequential/conv_lst_m2d/while/mul_4
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/mul_4:0 for ONNX tensor: sequential/conv_lst_m2d/while/mul_4:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/mul_4 [Mul] outputs: [sequential/conv_lst_m2d/while/mul_4:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/convolution_7__117 [Transpose]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/split_1:3
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/convolution_7__117 [Transpose] inputs: [sequential/conv_lst_m2d/while/split_1:3 -> (3, 3, 64, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/convolution_7__117 for ONNX node: sequential/conv_lst_m2d/while/convolution_7__117
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/convolution_7__117:0 for ONNX tensor: sequential/conv_lst_m2d/while/convolution_7__117:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/convolution_7__117 [Transpose] outputs: [sequential/conv_lst_m2d/while/convolution_7__117:0 -> (64, 64, 3, 3)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/convolution_7__115 [Transpose]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/mul_7:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/convolution_7__115 [Transpose] inputs: [sequential/conv_lst_m2d/while/mul_7:0 -> (-1, 32, 16, 64)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/convolution_7__115 for ONNX node: sequential/conv_lst_m2d/while/convolution_7__115
[03/21/2022-20:21:07] [V] [TRT] Registering tensor: sequential/conv_lst_m2d/while/convolution_7__115:0 for ONNX tensor: sequential/conv_lst_m2d/while/convolution_7__115:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/convolution_7__115 [Transpose] outputs: [sequential/conv_lst_m2d/while/convolution_7__115:0 -> (-1, 64, 32, 16)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Parsing node: sequential/conv_lst_m2d/while/convolution_7 [Conv]
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/convolution_7__115:0
[03/21/2022-20:21:07] [V] [TRT] Searching for input: sequential/conv_lst_m2d/while/convolution_7__117:0
[03/21/2022-20:21:07] [V] [TRT] sequential/conv_lst_m2d/while/convolution_7 [Conv] inputs: [sequential/conv_lst_m2d/while/convolution_7__115:0 -> (-1, 64, 32, 16)[FLOAT]], [sequential/conv_lst_m2d/while/convolution_7__117:0 -> (64, 64, 3, 3)[FLOAT]],
[03/21/2022-20:21:07] [V] [TRT] Convolution input dimensions: (-1, 64, 32, 16)
[03/21/2022-20:21:07] [V] [TRT] Kernel weights are not set yet. Kernel weights must be set using setInput(1, kernel_tensor) API call.
[03/21/2022-20:21:07] [V] [TRT] Registering layer: sequential/conv_lst_m2d/while/convolution_7 for ONNX node: sequential/conv_lst_m2d/while/convolution_7
[03/21/2022-20:21:07] [E] Error[4]: sequential/conv_lst_m2d/while/convolution_7: two inputs (data and weights) are allowed only in explicit-quantization mode.
[03/21/2022-20:21:07] [E] [TRT] ModelImporter.cpp:773: While parsing node number 13 [Loop -> "sequential/conv_lst_m2d/while_loop:0"]:
[03/21/2022-20:21:07] [E] [TRT] ModelImporter.cpp:774: --- Begin node ---
[03/21/2022-20:21:07] [E] [TRT] ModelImporter.cpp:775: input: "sequential/conv_lst_m2d_1/while/maximum_iterations:0"
input: "sequential_conv_lst_m2d_while_cond_frozen_2152_sequential/conv_lst_m2d/while/Less:0"
input: "sequential/conv_lst_m2d/time:0"
input: "Transpose__445:0"
input: "Transpose__445:0"
input: "sequential/conv_lst_m2d/strided_slice__48"
input: "sequential/conv_lst_m2d/split/ReadVariableOp/resource:0"
input: "sequential/conv_lst_m2d/split_1/ReadVariableOp/resource:0"
input: "sequential/conv_lst_m2d/split_2/ReadVariableOp/resource:0"
output: "sequential/conv_lst_m2d/while_loop:0"
output: "sequential/conv_lst_m2d/while_loop:1"
output: "sequential/conv_lst_m2d/while_loop:2"
output: "sequential/conv_lst_m2d/while_loop:3"
output: "sequential/conv_lst_m2d/while_loop:4"
output: "sequential/conv_lst_m2d/while_loop:5"
output: "sequential/conv_lst_m2d/while_loop:6"
output: "sequential/conv_lst_m2d/while_loop:7"
name: "sequential/conv_lst_m2d/while_loop"
...***(the describe of my onnx model node)***
[03/21/2022-16:54:55] [E] [TRT] ModelImporter.cpp:776: --- End node ---
[03/21/2022-16:54:55] [E] [TRT] ModelImporter.cpp:779: ERROR: ModelImporter.cpp:179 In function parseGraph:
[6] Invalid Node - sequential/conv_lst_m2d/while/convolution_7
sequential/conv_lst_m2d/while/convolution_7: two inputs (data and weights) are allowed only in explicit-quantization mode.
build() error!
double free or corruption (!prev)
When i google the solution of it, I found an analogous issue. May be here is the point:
[03/21/2022-20:21:07] [V] [TRT] Kernel weights are not set yet. Kernel weights must be set using setInput(1, kernel_tensor) API call.
But i dont know how to fix it.
python code
Here are my model code of python
model=Sequential()
model.add(Conv3D(filters=128,kernel_size=(11,11,1),strides=(4,4,1),padding='same',input_shape=(h,w,t_length,1),activation='tanh'))
model.add(Conv3D(filters=64,kernel_size=(5,5,1),strides=(2,2,1),padding='same',activation='tanh'))
model.add(ConvLSTM2D(filters=64,kernel_size=(3,3),strides=1,padding='same',dropout=0.4,recurrent_dropout=0.3,return_sequences=True))
model.add(ConvLSTM2D(filters=32,kernel_size=(3,3),strides=1,padding='same',dropout=0.3,return_sequences=True))
model.add(ConvLSTM2D(filters=64,kernel_size=(3,3),strides=1,return_sequences=True, padding='same',dropout=0.5))
model.add(Conv3DTranspose(filters=128,kernel_size=(5,5,1),strides=(2,2,1),padding='same',activation='tanh'))
model.add(Conv3DTranspose(filters=1,kernel_size=(11,11,1),strides=(4,4,1),padding='same',activation='tanh'))
model.compile(optimizer='adam',loss='mean_squared_error',metrics=['accuracy'])
#.h5 convert to .onnx
import tf2onnx
spec = (
tf.TensorSpec(
(None, args.h, args.w, args.t_length, 1),
tf.float32,
name=args.input_tensor_name,
),
)
model_proto, _ = tf2onnx.convert.from_keras(
model, input_signature=spec, output_path=onnx_output_path, opset=args.opset
)
Any advice is greatly appreciated!