Open hademircii opened 4 months ago
same issue. is there any known workaround?
a simple workaround is to work with a previous commit to main, right before #1274 ..
same here. trying this link.
Could you share the tensorrt_llm version you use? I try to reproduce but fail.
bhsueh@xxx:/tmp/TensorRT-LLM$ python3
Python 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from tensorrt_llm.lora_manager import LoraManager
[TensorRT-LLM] TensorRT-LLM version: 0.9.0.dev2024031200
I'm using this version.
>>> import tensorrt_llm
No protocol specified
[TensorRT-LLM] TensorRT-LLM version: 0.8.0
I think I should upgrade to latest and take a try.
I had this issue when was using 0.9.0 version installed from pip over 0.8.0 version of this repo. Updating repo to 0.9.0 fixed it.
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 15 days."
System Info
started receiving this error after rebuilding image today - that I believe picked up #1274 . (using
pip3 install tensorrt_llm -U --pre --extra-index-url https://pypi.nvidia.com
).Who can help?
kaiyux
Information
Tasks
examples
folder (such as GLUE/SQuAD, ...)Reproduction
follow the readme here https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/enc_dec/README.md#run to run line: https://github.com/NVIDIA/TensorRT-LLM/blob/4bb65f216f2f1dbddb022f4fdd8925c2856baa58/examples/enc_dec/run.py#L32
Expected behavior
import sucessfull.
actual behavior
import fails.
additional notes
-