depth_anything_vits14.onnx error :Dynamic dimensions required for input: image, but no shapes were provided. Automatically overriding shape to: 1x3x1x1

[zn123]

&&&& FAILED TensorRT.trtexec [TensorRT v100001] # trtexec.exe --onnx=E:/temp_python_ai/d3d_depth-anything-tensorrt/weights/depth_anything_vits14.onnx --saveEngine=depth_anything_vits14.engine --fp16

E:\temp_python_ai\d3d_depth-anything-tensorrt\External\TensorRT-10.0.1.6.Windows10.win10.cuda-11.8\TensorRT-10.0.1.6\bin>trtexec.exe --onnx=E:/temp_python_ai/d3d_depth-anything-tensorrt/weights/depth_anything_vits14.onnx --saveEngine=depth_anything_vits14.engine --fp16
&&&& RUNNING TensorRT.trtexec [TensorRT v100001] # trtexec.exe --onnx=E:/temp_python_ai/d3d_depth-anything-tensorrt/weights/depth_anything_vits14.onnx --saveEngine=depth_anything_vits14.engine --fp16
[11/22/2024-14:39:04] [I] === Model Options ===
[11/22/2024-14:39:04] [I] Format: ONNX
[11/22/2024-14:39:04] [I] Model: E:/temp_python_ai/d3d_depth-anything-tensorrt/weights/depth_anything_vits14.onnx
[11/22/2024-14:39:04] [I] Output:
[11/22/2024-14:39:04] [I] === Build Options ===
[11/22/2024-14:39:04] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default, tacticSharedMem: default
[11/22/2024-14:39:04] [I] avgTiming: 8
[11/22/2024-14:39:04] [I] Precision: FP32+FP16
[11/22/2024-14:39:04] [I] LayerPrecisions:
[11/22/2024-14:39:04] [I] Layer Device Types:
[11/22/2024-14:39:04] [I] Calibration:
[11/22/2024-14:39:04] [I] Refit: Disabled
[11/22/2024-14:39:04] [I] Strip weights: Disabled
[11/22/2024-14:39:04] [I] Version Compatible: Disabled
[11/22/2024-14:39:04] [I] ONNX Plugin InstanceNorm: Disabled
[11/22/2024-14:39:04] [I] TensorRT runtime: full
[11/22/2024-14:39:04] [I] Lean DLL Path:
[11/22/2024-14:39:04] [I] Tempfile Controls: { in_memory: allow, temporary: allow }
[11/22/2024-14:39:04] [I] Exclude Lean Runtime: Disabled
[11/22/2024-14:39:04] [I] Sparsity: Disabled
[11/22/2024-14:39:04] [I] Safe mode: Disabled
[11/22/2024-14:39:04] [I] Build DLA standalone loadable: Disabled
[11/22/2024-14:39:04] [I] Allow GPU fallback for DLA: Disabled
[11/22/2024-14:39:04] [I] DirectIO mode: Disabled
[11/22/2024-14:39:04] [I] Restricted mode: Disabled
[11/22/2024-14:39:04] [I] Skip inference: Disabled
[11/22/2024-14:39:04] [I] Save engine: depth_anything_vits14.engine
[11/22/2024-14:39:04] [I] Load engine:
[11/22/2024-14:39:04] [I] Profiling verbosity: 0
[11/22/2024-14:39:04] [I] Tactic sources: Using default tactic sources
[11/22/2024-14:39:04] [I] timingCacheMode: local
[11/22/2024-14:39:04] [I] timingCacheFile:
[11/22/2024-14:39:04] [I] Enable Compilation Cache: Enabled
[11/22/2024-14:39:04] [I] errorOnTimingCacheMiss: Disabled
[11/22/2024-14:39:04] [I] Preview Features: Use default preview flags.
[11/22/2024-14:39:04] [I] MaxAuxStreams: -1
[11/22/2024-14:39:04] [I] BuilderOptimizationLevel: -1
[11/22/2024-14:39:04] [I] Calibration Profile Index: 0
[11/22/2024-14:39:04] [I] Weight Streaming: Disabled
[11/22/2024-14:39:04] [I] Debug Tensors:
[11/22/2024-14:39:04] [I] Input(s)s format: fp32:CHW
[11/22/2024-14:39:04] [I] Output(s)s format: fp32:CHW
[11/22/2024-14:39:04] [I] Input build shapes: model
[11/22/2024-14:39:04] [I] Input calibration shapes: model
[11/22/2024-14:39:04] [I] === System Options ===
[11/22/2024-14:39:04] [I] Device: 0
[11/22/2024-14:39:04] [I] DLACore:
[11/22/2024-14:39:04] [I] Plugins:
[11/22/2024-14:39:04] [I] setPluginsToSerialize:
[11/22/2024-14:39:04] [I] dynamicPlugins:
[11/22/2024-14:39:04] [I] ignoreParsedPluginLibs: 0
[11/22/2024-14:39:04] [I]
[11/22/2024-14:39:04] [I] === Inference Options ===
[11/22/2024-14:39:04] [I] Batch: Explicit
[11/22/2024-14:39:04] [I] Input inference shapes: model
[11/22/2024-14:39:04] [I] Iterations: 10
[11/22/2024-14:39:04] [I] Duration: 3s (+ 200ms warm up)
[11/22/2024-14:39:04] [I] Sleep time: 0ms
[11/22/2024-14:39:04] [I] Idle time: 0ms
[11/22/2024-14:39:04] [I] Inference Streams: 1
[11/22/2024-14:39:04] [I] ExposeDMA: Disabled
[11/22/2024-14:39:04] [I] Data transfers: Enabled
[11/22/2024-14:39:04] [I] Spin-wait: Disabled
[11/22/2024-14:39:04] [I] Multithreading: Disabled
[11/22/2024-14:39:04] [I] CUDA Graph: Disabled
[11/22/2024-14:39:04] [I] Separate profiling: Disabled
[11/22/2024-14:39:04] [I] Time Deserialize: Disabled
[11/22/2024-14:39:04] [I] Time Refit: Disabled
[11/22/2024-14:39:04] [I] NVTX verbosity: 0
[11/22/2024-14:39:04] [I] Persistent Cache Ratio: 0
[11/22/2024-14:39:04] [I] Optimization Profile Index: 0
[11/22/2024-14:39:04] [I] Weight Streaming Budget: Disabled
[11/22/2024-14:39:04] [I] Inputs:
[11/22/2024-14:39:04] [I] Debug Tensor Save Destinations:
[11/22/2024-14:39:04] [I] === Reporting Options ===
[11/22/2024-14:39:04] [I] Verbose: Disabled
[11/22/2024-14:39:04] [I] Averages: 10 inferences
[11/22/2024-14:39:04] [I] Percentiles: 90,95,99
[11/22/2024-14:39:04] [I] Dump refittable layers:Disabled
[11/22/2024-14:39:04] [I] Dump output: Disabled
[11/22/2024-14:39:04] [I] Profile: Disabled
[11/22/2024-14:39:04] [I] Export timing to JSON file:
[11/22/2024-14:39:04] [I] Export output to JSON file:
[11/22/2024-14:39:04] [I] Export profile to JSON file:
[11/22/2024-14:39:04] [I]
[11/22/2024-14:39:04] [I] === Device Information ===
[11/22/2024-14:39:04] [I] Available Devices:
[11/22/2024-14:39:04] [I] Device 0: "NVIDIA GeForce RTX 3090" UUID: GPU-a672ed59-b38b-993f-86e0-e1b4683f56ca
[11/22/2024-14:39:04] [I] Selected Device: NVIDIA GeForce RTX 3090
[11/22/2024-14:39:04] [I] Selected Device ID: 0
[11/22/2024-14:39:04] [I] Selected Device UUID: GPU-a672ed59-b38b-993f-86e0-e1b4683f56ca
[11/22/2024-14:39:04] [I] Compute Capability: 8.6
[11/22/2024-14:39:04] [I] SMs: 82
[11/22/2024-14:39:04] [I] Device Global Memory: 24575 MiB
[11/22/2024-14:39:04] [I] Shared Memory per SM: 100 KiB
[11/22/2024-14:39:04] [I] Memory Bus Width: 384 bits (ECC disabled)
[11/22/2024-14:39:04] [I] Application Compute Clock Rate: 1.695 GHz
[11/22/2024-14:39:04] [I] Application Memory Clock Rate: 9.751 GHz
[11/22/2024-14:39:04] [I]
[11/22/2024-14:39:04] [I] Note: The application clock rates do not reflect the actual clock rates that the GPU is currently running at.
[11/22/2024-14:39:04] [I]
[11/22/2024-14:39:04] [I] TensorRT version: 10.0.1
[11/22/2024-14:39:04] [I] Loading standard plugins
[11/22/2024-14:39:05] [I] [TRT] [MemUsageChange] Init CUDA: CPU +73, GPU +0, now: CPU 17983, GPU 1269 (MiB)
[11/22/2024-14:39:07] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +2654, GPU +310, now: CPU 20917, GPU 1579 (MiB)
[11/22/2024-14:39:07] [I] Start parsing network model.
[11/22/2024-14:39:07] [I] [TRT] ----------------------------------------------------------------
[11/22/2024-14:39:07] [I] [TRT] Input filename: E:/temp_python_ai/d3d_depth-anything-tensorrt/weights/depth_anything_vits14.onnx
[11/22/2024-14:39:07] [I] [TRT] ONNX IR version: 0.0.8
[11/22/2024-14:39:07] [I] [TRT] Opset version: 17
[11/22/2024-14:39:07] [I] [TRT] Producer name: pytorch
[11/22/2024-14:39:07] [I] [TRT] Producer version: 2.1.2
[11/22/2024-14:39:07] [I] [TRT] Domain:
[11/22/2024-14:39:07] [I] [TRT] Model version: 0
[11/22/2024-14:39:07] [I] [TRT] Doc string:
[11/22/2024-14:39:07] [I] [TRT] ----------------------------------------------------------------
[11/22/2024-14:39:07] [I] Finished parsing network model. Parse time: 0.1201
[11/22/2024-14:39:07] [W] Dynamic dimensions required for input: image, but no shapes were provided. Automatically overriding shape to: 1x3x1x1
[11/22/2024-14:39:07] [I] Set shape of input tensor image for optimization profile 0 to: MIN=1x3x1x1 OPT=1x3x1x1 MAX=1x3x1x1
[11/22/2024-14:39:07] [E] Error[4]: [graphShapeAnalyzer.cpp::nvinfer1::builder::`anonymous-namespace'::ShapeAnalyzerImpl::analyzeShapes::2084] Error Code 4: Miscellaneous (IResizeLayer /Resize_1: /Resize_1: IResizeLayer requires that if input dimension is zero, output dimension must be zero too (axis = 2 input dimension = 0 output dimension = 1)
)
[11/22/2024-14:39:07] [E] Engine could not be created from network
[11/22/2024-14:39:07] [E] Building engine failed
[11/22/2024-14:39:07] [E] Failed to create engine from model or file.
[11/22/2024-14:39:07] [E] Engine set up failed
&&&& FAILED TensorRT.trtexec [TensorRT v100001] # trtexec.exe --onnx=E:/temp_python_ai/d3d_depth-anything-tensorrt/weights/depth_anything_vits14.onnx --saveEngine=depth_anything_vits14.engine --fp16

[zn123]

Reading this [https://github.com//issues/26] doesn't solve my problem
my cuda is:
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8+cudnn-windows-x86_64-8.9.7.29_cuda11-archive

[zn123]

onnx from:
https://github.com/fabio-sim/Depth-Anything-ONNX/releases

[zn123]

i see
[https://github.com/fabio-sim/Depth-Anything-ONNX/releases] cannot be used for is