Closed (tensorml) user@10324cdf-227b-465d-98af-cabfa0a5996c:~/TensorRT-LLM$ python3 ./scripts/build_wheel.py --trt_root /home/user/TensorRT-LLM/TensorRT-9.1.0.4 Lookinginindexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com Requirement already satisfied: buildin/home/user/mamb...
As I can't upgrade my CUDA version, I am trying to build onnxruntime v1.8.1, as I understand it should work fine with CUDA 10.2. The only way I manage to download an onnxruntime version that seem close to 1.8.1 was by switching to the rel-1.8.1 branch on this repo. But afte...
self.tokenizer = AutoTokenizer.from_pretrained(tokenizer_name_or_path, **hf_tokenizer_kwargs) File "/usr/local/lib/python3.9/site-packages/transformers/models/auto/tokenization_auto.py", line 736, in from_pretrained return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, ...
I am adapting a large language model for NPU. Since the opt model is only a basic model and cannot complete the dialogue function, I have to adapt other models. I have followed code: @echo off python run_onnx.py --model_name %1 --onnx op...
Issue I got the above compiling error when trying to install the git pydensecrf part. I have ensured the Win11 SDK is installed (not the Win10 one - that was installed earlier but didnt change the error), cython is installed locally as w...
Cloninghttps://github.com/IDEA-Research/GroundingDINOto c:\windows\temp\pip-req-build-27jqfcl_ Resolvedhttps://github.com/IDEA-Research/GroundingDINOto commit 57535c5a79791cb76e36fdb64975271354f10251 Preparing metadata (setup.py): started
🐛 Describe the bug I meet the segment fault when use Megatron-LM firstly, and then I reproduce it using simple torch code. import torch import os os.environ['MASTER_ADDR']='{my_ip}' os.environ['MASTER_PORT']='6001' torch.distributed.init...