lnccbrown / lnccbrown/LANfactory

Extend network inspection

Open
#100 3 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
16
Forks
4
Avg merge
3d 7h
Merged PRs (30d)
8

Description

After migrating the core features from hddm in #82 and #99, network inspection remains notebook-only. We hope to add a first-class inspection workflow with discovery of pretrained models.

## Summary
Network inspection in LANfactory is currently notebook-bound and effectively limited to a single interface. The current path depends on manual notebook execution and requires building models or generating data during runtime before inspection can happen.

## Problem

- Inspection is tied to a notebook-driven flow instead of a supported product workflow.
- Users must manage paths and runtime setup manually.
- Model discovery is missing as a first-class capability.

## Proposed Outcome

- Introduce a first-class inspection workflow outside notebook-only usage.
- Support selecting and inspecting existing trained models directly.
- Keep notebook support as supplemental, not primary.

## Scope

- Workflow entry point via documented command.
- Artifact discovery across common folder layouts, including nested network-type directories.
- Inspection views for KDE-vs-LAN and manifold once a valid model is selected.

## Acceptance Criteria

- User can launch inspection from a documented command.
- Newly trained models are auto-discovered without manual deep-path entry.
- Invalid or incomplete model folders produce clear errors.
- Notebook tutorial remains available.

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the existing notebook-driven inspection flow and the current model folder layouts. Identify the entry point for a documented inspection command, then trace how KDE-vs-LAN and manifold views are selected. Done means existing trained models are discovered without deep manual paths, invalid folders produce clear errors, and the notebook tutorial remains available.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
cli, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.