# ImagineAILab/ai-by-hand-excel

Repository: https://github.com/ImagineAILab/ai-by-hand-excel
Canonical: https://ross.abutalabs.com/products/ai-by-hand-excel
License: MIT
License Family: permissive
Last push: 2025-01-28T13:33:23+00:00

## Health v2 (maintenance only)
Score: 24/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 3, release rhythm 35, longevity 51
- inputs: {"age_days": 717, "days_push": 582, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6261, forks 779 (observed 2026-08-28T04:09:41.108380+00:00)

## What it is
A collection of Excel workbooks that implement deep learning concepts by hand, from softmax and backpropagation to transformers, LSTM, Mamba, and DeepSeek architectures. It is an educational resource for learning neural network mechanics visually through spreadsheet formulas rather than code.

## Use cases
- learn how backpropagation works step by step
- understand self-attention and multihead attention in a spreadsheet
- study LSTM and xLSTM cell mechanics visually
- teach a deep learning class with hands-on Excel exercises
- explore how the transformer architecture works without writing code
- work through softmax and temperature scaling by hand

## When to choose
- you want to deeply understand neural network math by building it manually in Excel
- you are an instructor looking for visual teaching materials for deep learning
- you prefer spreadsheet-based learning over coding exercises

## When to avoid
- you need production-ready machine learning code or libraries
- you want runnable training pipelines or GPU acceleration
- you do not have access to Microsoft Excel or a compatible spreadsheet application

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, deep-learning, data-visualization
- domain: machine-learning, deep-learning, education, tutorials
- platform: windows, cross-platform
- tags: excel, spreadsheets, hands-on-learning, neural-networks, transformers, attention, lstm, backpropagation, deepseek, mamba, education, macos

## Member repositories
- ImagineAILab/ai-by-hand-excel (main) score 24

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:41.108380+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:46:30.722108+00:00, confidence not recorded.
  - readme: https://github.com/ImagineAILab/ai-by-hand-excel (fetched 2026-08-28T04:09:41.108380+00:00, sha ff5612961687)
- Data as of 2026-08-30T08:39:29.467469+00:00.