[devcontainer] TensorRT provider fails with missing libnvinfer.so.10 in focoos-tensorrt image

[AlfonsoCorrado]

I’m using the dev container with the tag: Focoos GPU (OnnxRuntime, TensorRT), and encountered an error when trying to export a model to ONNX_TRT16. The runtime fails with the following error:

[E:onnxruntime:Default, provider_bridge_ort.cc:2167 TryGetProviderInfo_TensorRT] /onnxruntime_src/onnxruntime/core/session/provider_bridge_ort.cc:1778 onnxruntime::Provider& onnxruntime::ProviderLibrary::Get() [ONNXRuntimeError] : 1 : FAIL : Failed to load library libonnxruntime_providers_tensorrt.so with error: libnvinfer.so.10: cannot open shared object file: No such file or directory

Full error also includes:

EP Error /onnxruntime_src/onnxruntime/python/onnxruntime_pybind_state.cc:505 void onnxruntime::python::RegisterTensorRTPluginsAsCustomOps(PySessionOptions&, const onnxruntime::ProviderOptions&) Please install TensorRT libraries as mentioned in the GPU requirements page, make sure they're in the PATH or LD_LIBRARY_PATH, and that your GPU is supported.
when using [('TensorrtExecutionProvider', {'device_id': 0, 'trt_fp16_enable': True, 'trt_force_sequential_engine_build': False, 'trt_engine_cache_enable': True, 'trt_engine_cache_path': '/root/FocoosAI/models/f6bc95750ba6453e/.trt_cache', 'trt_ep_context_file_path': '/root/FocoosAI/models/f6bc95750ba6453e', 'trt_timing_cache_enable': True, 'trt_builder_optimization_level': 3, 'trt_timing_cache_path': '/root/FocoosAI/models/.trt_timing_cache'}), ('CPUExecutionProvider', {})]
Falling back to ['CPUExecutionProvider'] and retrying.

It appears that the required TensorRT libraries are missing or not linked properly inside the container.

Steps to Reproduce:
1. Use the Focoos GPU (OnnxRuntime, TensorRT) dev container
2. Attempt to export a model with ONNX_TRT16 (FP16)
3. Observe the failure and fallback to CPUExecutionProvider

[AlfonsoCorrado]

Could be solved adding this in the Dockerfile:

ENV LD_LIBRARY_PATH=/usr/local/lib/python3.12/dist-packages/tensorrt_libs:${LD_LIBRARY_PATH}