Upgraded DeviceMonitor

[kozlov721]

Purpose

Allows to monitor more metrics

Specification

• Added RAM and CPU monitoring
• Temperature monitoring
• DSP frequencies instead of just the overall utilization
• Merged all the monitors into a single DeviceMonitor

Dependencies & Potential Impact

None / not applicable

Deployment Plan

None / not applicable

Testing & Validation

None / not applicable

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: dd6d40b5e6

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[ptoupas]

A couple extra comments on top of the on-code comments I provided:

1. Setting just the --device-id ... flag on a benchmark run (e.g. modelconverter benchmark rvc4 --model-path luxonis/yolov8-nano-pose-estimation:coco-512x288 --power-benchmark --dsp-benchmark --device-id da79b2bf) throws the following warning: WARNING No device ID specified, using the first connected device: bb81a24e. which shouldn't be an issue given that we are providing the exact device-id. On the other hand if just the --device-ip is provided the --device-id is inferred and found properly
2. It seems like if any of the --power-benchmark, --dsp-benchmark flags are provided in CLI the other one is also enabled by default. Even if I explicitly set the other one to --no-... the respective metric is tracked while it shouldn't. For example the following flags track both power and dsp --power-benchmark --no-dsp-benchmark. Same if I just set one like --power-benchmark or --dsp-benchmark both are tracked while it shouldn't.