Neural-Lab / README.md
KrazyKitty's picture
Convert NEURAL LAB to verified Gradio research dashboard
69458c4 verified
|
Raw History Blame Contribute Delete
1.99 kB
metadata
title: KRAZYKITTY // NEURAL LAB
emoji: 🐱
colorFrom: purple
colorTo: purple
sdk: gradio
sdk_version: 6.30.0
python_version: 3.12
app_file: app.py
pinned: false
license: apache-2.0
short_description: KrazyKitty's ultraviolet AI research dashboard
fullWidth: true

KRAZYKITTY // NEURAL LAB

NEURAL LAB is the standalone Gradio dashboard for KrazyKitty's Hugging Face research portfolio. It links KittyLM, DATA LAB, NEKO//AGENT, curated research Collections, and future demos through a consistent ultraviolet interface.

Current behaviour

  • Presents interactive project and Collection explorers with visible evidence states.
  • Links to the verified public KittyLM, DATA LAB, and NEKO//AGENT research repositories.
  • Links directly to all four public Research Collections.
  • Includes a deterministic research-readiness checklist for language-model, dataset, and agent-system work.
  • Distinguishes published artifacts, planned work, limitations, and approval gates in visible text.
  • Loads no model, dataset, retrieval index, telemetry service, or external API.
  • Does not present generated text or canned content as model output.
  • Does not certify that an experiment is safe, valid, or ready to publish.

Local preview

Install the pinned Gradio version and run the app, for example:

python -m pip install gradio==6.30.0
python app.py

Accessibility and responsive design

  • labelled interactive project, Collection, and checklist controls;
  • semantic headings, landmarks, and external-link descriptions;
  • visible keyboard focus;
  • palette contrast designed around the approved violet scale;
  • single-column layouts at narrow widths;
  • reduced-motion support;
  • no colour-only status labels;
  • readable line lengths and scalable type.

Publication status

This public Gradio Space is a research navigation and documentation tool. It is not a live model demo, agent, benchmark service, or evidence that any planned model has been trained or released.