# oreilly-japan/deep-learning-from-scratch-2

『ゼロから作る Deep Learning ❷』(O'Reilly Japan, 2018)

Repository: https://github.com/oreilly-japan/deep-learning-from-scratch-2
Canonical: https://ross.abutalabs.com/products/deep-learning-from-scratch-2
Language: Python
License: MIT
License Family: permissive
Last push: 2024-08-19T03:06:33+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3117, "days_push": 744, "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 1249, forks 794 (observed 2026-08-28T04:04:07.913687+00:00)

## What it is
Companion source code repository for the O'Reilly Japan book 'Deep Learning from Scratch 2: Natural Language Processing' (2018). It contains per-chapter Python/NumPy implementations of NLP deep learning techniques such as word2vec, RNN/LSTM language models, and attention mechanisms.

## Use cases
- learn deep learning for NLP from scratch
- implement word2vec without frameworks
- understand RNN and LSTM internals
- study attention mechanism code
- follow a deep learning textbook with runnable code

## When to choose
- you are reading the book and want its official code
- you want framework-free NumPy implementations to learn fundamentals
- you prefer minimal dependencies (NumPy/Matplotlib)

## When to avoid
- you need a production-ready NLP or deep learning library
- you want GPU-optimized training at scale (CuPy is optional and limited)
- you need maintained, feature-rich tooling rather than educational code

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: deep-learning, nlp, machine-learning
- domain: deep-learning, education, tutorials
- platform: python, cross-platform
- tags: book-companion-code, word2vec, rnn, attention, numpy-only, japanese-book, natural-language-processing

## Member repositories
- oreilly-japan/deep-learning-from-scratch-2 (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:07.913687+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 2026-08-30T05:07:48.420060+00:00, confidence not recorded.
  - readme: https://github.com/oreilly-japan/deep-learning-from-scratch-2 (fetched 2026-08-28T04:04:07.913687+00:00, sha 51ec4a927205)
- Data as of 2026-08-30T08:39:29.467469+00:00.