[Model/Framework] error in run the script run_inference_on_triton.py

[sepzjh]

Related to Model/Framework(s)
Triton
Describe the bug
A clear and concise description of what the bug is.
when i run the scipt:
python3 triton/run_inference_on_triton.py
--server-url localhost:8001
--model-name ${MODEL_NAME}
--model-version 1
--output-dir ${SHARED_DIR}/accuracy_dump

--precision ${PRECISION}
--dataloader triton/dataloader.py
--data-dir ${DATASETS_DIR}/imagenet
--batch-size ${MAX_BATCH_SIZE}
--dump-labels
error:
2024-05-06 08:36:19,172 INFO dataloader Creating PIL dataloader on data_dir=/data/DeepLearningExamples/PyTorch/Classification/ConvNets/workspace/datasets_dir/imagenet #images=50000 image_size=(224, 224) batch_size=16
Exception in thread Thread-1:
Traceback (most recent call last):
File "/opt/conda/lib/python3.8/threading.py", line 932, in _bootstrap_inner
self.run()
File "/opt/conda/lib/python3.8/threading.py", line 870, in run
self._target(*self._args, **self._kwargs)
File "triton/run_inference_on_triton.py", line 165, in req_loop
infer_input.set_data_from_numpy(data)
File "/opt/conda/lib/python3.8/site-packages/tritonclient/grpc/init.py", line 1441, in set_data_from_numpy
self._input.parameters.pop('shared_memory_region', None)
TypeError: pop() takes from 1 to 2 positional arguments but 3 were given

To Reproduce

follow the Guide:
DeepLearningExamples/PyTorch/Classification/ConvNets/triton/resnet50
/README.md Quick Start Guide

model_store

• resnet50
  • config.pbtxt
  • 1
    • model.onnx

Expected behavior
no error in running the scirpt and have accuracy_metrics.csv

Environment
Please provide at least:

• Container version: nvcr.io/nvidia/tritonserver:21.02-py3 and resnet50:latest
• GPUs in the system: 1x NVIDIA A800 80GB PCIe):
• CUDA driver version :525.85.12
• triton_version:2.7.0