[Docs] Difference between profiler and deploy script

[Daanfb]

📚 The doc issue

I want to know the difference between tools/profiler.py and tools/deploy.py script.

I know that with tools/deploy.py script you can get the model in ONNX, TensorRT, etc, but you can use that backends with just tools/profiler.py?

What's the difference between export the model to one of that backend with deploy script and then write the code to infer it, and just run the profile script?

I mean, with this command:

python tools/profiler.py configs/mmpose/pose-detection_rtmo_tensorrt-fp16_dynamic-640x640.py ../mmpose/configs/body_2d_keypoint/rtmo/body7/rtmo-s_8xb32-600e_body7-640x640.py C:/Users/Jordi/Downloads/val2017 --model https://download.openmmlab.com/mmpose/v1/projects/rtmo/rtmo-s_8xb32-600e_body7-640x640-dac2bf74_20231211.pth --shape 640x640 --device cuda --num-iter 100 --warmup 10 --batch-size 8

Am I using the model in TensorRT?

Suggest a potential alternative/fix

No response