search

triton-inference-server / model_navigator

Triton Model Navigator is an inference toolkit designed for optimizing and deploying Deep Learning models with a focus on NVIDIA GPUs.
https://triton-inference-server.github.io/model_navigator/
Apache License 2.0
176 stars 24 forks source link

Readme update #30

Open esnvidia opened 10 months ago

esnvidia commented 10 months ago

This install: pip install -U --extra-index-url https://pypi.ngc.nvidia.com triton-model-navigator[<extras,>]

breaks pips parser.

This works instead:

pip install -U --extra-index-url https://pypi.ngc.nvidia.com triton-model-navigator[extras]

jkosek commented 10 months ago

Hi @esnvidia!

The command you are mentioning is just an example how to use extras during pip installation.

We may need to enhance the instruction to be more clear. To give an example the following installation commands could be used based on model framework:

  1. PyTorch 
    pip install -U --extra-index-url https://pypi.ngc.nvidia.com triton-model-navigator
  2. TensorFlow 
    pip install -U --extra-index-url https://pypi.ngc.nvidia.com triton-model-navigator[tensorflow]
  3. JAX 
    pip install -U --extra-index-url https://pypi.ngc.nvidia.com triton-model-navigator[jax]

Let us know if this explain reported problem. Thanks!

esnvidia commented 10 months ago

Hi @jkosek! Yes, it does, however, I should have been more clear in my request. I simply meant to update the readme to be clearer on the options and ensure that it is able to be copied direct to the terminal.

jkosek commented 10 months ago

This is great feedback! Thanks for pointing that @esnvidia.

github-actions[bot] commented 8 months ago

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 7 days.

jkosek commented 5 months ago

The docs have been improved. Now the installation is described as mentioned above in the issue.