• Hardware (T4/V100/Xavier/Nano/etc) T4
• Network Type (Detectnet_v2/Faster_rcnn/Yolo_v4/LPRnet/Mask_rcnn/Classification/etc) Emotionnet
• TLT Version (Please run “tlt info --verbose” and share “docker_tag” here) tao-converter_v3.22.05_trt8.4_x86.zip, and I have TensorRT 8.4.3-1 and CUDA11.6 installed.
• Training spec file(If have, please share here)
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)
Hello, I am trying to use tao-converter in order to serialise the Emotionnet model downloaded from here: EmotionNet | NVIDIA NGC
But I am getting an error asking me to provide an optimization profile, and I don’t know what to use for that option.
[134]build:~/volume $ ./tao-converter -k ${key} model.etlt -m 16 -b 8 -i nchw -d 1,136,1 -e saved.engine -t int8
[INFO] [MemUsageChange] Init CUDA: CPU +314, GPU +0, now: CPU 321, GPU 241 (MiB)
[INFO] [MemUsageChange] Init builder kernel library: CPU +207, GPU +68, now: CPU 545, GPU 309 (MiB)
[INFO] ----------------------------------------------------------------
[INFO] Input filename: /tmp/fileH2OrU6
[INFO] ONNX IR version: 0.0.5
[INFO] Opset version: 10
[INFO] Producer name: tf2onnx
[INFO] Producer version: 1.6.3
[INFO] Domain:
[INFO] Model version: 0
[INFO] Doc string:
[INFO] ----------------------------------------------------------------
[WARNING] onnx2trt_utils.cpp:369: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[INFO] Detected input dimensions from the model: (-1, 1, 136, 1)
[ERROR] Model has dynamic shape but no optimization profile specified.
Could you help me please? Thank you