-
Notifications
You must be signed in to change notification settings - Fork 349
Replies: 1 comment · 9 replies
-
Our main branch should be compatible with TRT 8.5.2. Can you try compiling it with TRT 8.5.2 and try ? |
Beta Was this translation helpful? Give feedback.
All reactions
-
Just in case I hadn't provided enough context with the previous post of the error here is the extended traceback:
|
Beta Was this translation helpful? Give feedback.
All reactions
-
torch_tensorrt.compile under the hood uses torch.jit.script for the sake of correctness so there arent models that trace but produce erroneous results. You can provide a traced model to torch_tensorrt.compile if you already have a working torchscript model |
Beta Was this translation helpful? Give feedback.
All reactions
-
Changing the code to:
Produces the following error:
And adding the truncate_long_and_double argument like so:
Produces no error but actually runs slower than base TorchScript |
Beta Was this translation helpful? Give feedback.
All reactions
-
If the model is not compiled fully, i.e not all operations are run in tensorrt then perf might be lower for instance if there are a lot of context switches from pytorch to trt. Settings like min block size, strategic additions of converters or assigning certain ops or modules to run in pytorch can help you tune the performance of the compiled model |
Beta Was this translation helpful? Give feedback.
All reactions
-
Is there something I'm doing wrong so that it doesn't compile fully? |
Beta Was this translation helpful? Give feedback.
-
Sorry if this has already been asked or answered elsewhere, but are there plans to implement support for TensorRT 8.5.2? I am using the Segformer model and it is not supported in TensorRT 8.5 but is supported by 8.5.2. If there are plans already, is there a timeline for when that would be implemented? Even if the timeline is fluid or merely an estimation it would be helpful. Thank you!
Beta Was this translation helpful? Give feedback.
All reactions