HFOnnx expects ORT provider

[chintanckg]

embeddings = onnx("sentence-transformers/paraphrase-MiniLM-L6-v2", "pooling", "embeddings.onnx", quantize=True)

ValueError: This ORT build has ['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider'] enabled. Since ORT 1.9, you are required to explicitly set the providers parameter when instantiating InferenceSession. For example, onnxruntime.InferenceSession(..., providers=['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider'], ...)

[davidmezzetti]

Thank you for reporting this issue! The problem comes in when onnxruntime-gpu is installed as that has multiple providers. Logic has been added to set providers based on the available providers.

This will be available when 4.0 is released. It can also be installed from source right now.

[chintanckg]

Thank you for reporting this issue! The problem comes in when onnxruntime-gpu is installed as that has multiple providers. Logic has been added to set providers based on the available providers.

This will be available when 4.0 is released. It can also be installed from source right now.

Thanks @davidmezzetti