Closed pfldy2850 closed 5 months ago
I get the exact same error when trying to install ammo
pip download --extra-index-url https://pypi.nvidia.com nvidia-ammo
I take a try but I cannot reproduce your issues
bhsueh@xxx:/home/scratch.bhsueh_sw_1$ nvidia-docker run -ti --gpus all --shm-size 25g nvcr.io/nvidia/tritonserver:24.02-trtllm-python-py3 bash
=============================
== Triton Inference Server ==
=============================
NVIDIA Release 24.02 (build 83572707)
Triton Server Version 2.43.0
Copyright (c) 2018-2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
Various files include modifications (c) NVIDIA CORPORATION & AFFILIATES. All rights reserved.
This container image and its contents are governed by the NVIDIA Deep Learning Container License.
By pulling and using the container, you accept the terms and conditions of this license:
https://developer.nvidia.com/ngc/nvidia-deep-learning-container-license
NOTE: CUDA Forward Compatibility mode ENABLED.
Using CUDA 12.3 driver version 545.23.08 with kernel driver version 535.129.03.
See https://docs.nvidia.com/deploy/cuda-compatibility/ for details.
Could you try again and share the full log?
Thank you for your response, @byshiue
I am now doubting the following log part of what I wrote.
Looking in indexes: https://pypi.org/simple, https://pypi.nvidia.com
WARNING: Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ProtocolError('Connection aborted.', ConnectionResetError(104, 'Connection reset by peer'))': /tensorrt-llm/
I'm guessing from those logs that the firewall or network settings in my environment are preventing normal access to the pypi index.
I get the same error when trying to install tensorrt-llm-0.8.0 in docker:
docker pull nvcr.io/nvidia/tritonserver:24.02-trtllm-python-py3
...
pip install tensorrt_llm-0.8.0-cp310-cp310-linux_x86_64.whl
Error:
Collecting nvidia-ammo~=0.7.0 (from tensorrt-llm==0.8.0)
Downloading nvidia-ammo-0.7.4.tar.gz (6.9 kB)
Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error
x python setup.py egg info did not run successfully.
exit code: 1
[6 lines of output]
Traceback (most recent call last):
File "<string>", line 2, in <module>
File "<pip-setuptools-caller>", line 34, in <module>
File "/tmp/pip-install-kif3gunq/nvidia-ammo_2f5a5762a60446e69eb1c0693b55ac14/setup.py", line 90, in <module>
raise RuntimeError("Bad params")
RuntimeError: Bad params
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed
Might you try running
pip install requirements-dev.txt
I ran into this as well and found the root cause I think. If someone runs into this again, issue was that 0.7.* versions of nvidia-ammo on pypi seem broken. It's important to add --extra-index-url https://pypi.nvidia.com
. The version on pypi.nvidia.com seem to be working.
System Info
Who can help?
@byshiue
Information
Tasks
examples
folder (such as GLUE/SQuAD, ...)Reproduction
I followed the official installation guide and ran the command as shown below.
Expected behavior
installation success
actual behavior
installation failed
additional notes
I used the official image, nvcr.io/nvidia/tritonserver:24.02-trtllm-python-py3, and the python version is 3.10.
What I found is that during the download from pypi, I get tensorrt-llm-0.8.0.tar.gz, which appears to be a postfix-less name for the platform and python version.