Kevin Chen comments

Results 88 comments of


                                            Kevin Chen

tensorrt use onnx_parser to parse onnx file error

Are you able the share the model?

Einsum_110: einsum input tensor 0 has invalid type Int32

This is a unfortunately a known limitation in the Einsum Layer in TRT - we only support floating-point types for Einsum equations. Do you know which operation this Einsum equation...

Einsum_110: einsum input tensor 0 has invalid type Int32

It's possible that one of the inputs are being interpreted incorrectly as INT32. Can you provide the converted .onnx model?

Parser does not provide the correct output dimension

What version of TRT are yo using? Are you able to share the model?

Problem converting ONNX model to TensorRT 7.2.1(dien.onnx)

It looks like you are using the opset 13 version of Unsqueeze which is currently unsupported. Can you try exporting your model to a lower opset (i.e. opset 11)?

Windows make-j issue

Which protobuf version are you using?

onnx2trt: No speed difference between models of different sizes.

What TRT version are you using? Are you able to provide the models you are benchmarking?

onnx2trt: No speed difference between models of different sizes.

Yes, proving the models in ONNX form will be useful. Are you seeing the same performance difference with the latest version of TRT?

how to use precision 16 in python module?

I recommend using the tools specified in https://github.com/onnx/onnx-tensorrt/blob/master/docs/faq.md#how-do-i-import-and-run-an-onnx-model-through-tensorrt for generating TensroRT engines from ONNX models since those tools are more supported.

BART Error: 'BARTTRTDecoder' object has no attribute 'trt_context_non_kv'

You can set `use_cache=False` for now. The kv cache feature is not fully supported in notebooks yet. We'll add updated notebooks supporting this feature in one of our next releases.