iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
my_public_notebooks is described by its author, Vitor Calvi, as a collection of Jupyter notebooks for Google Colab and local GPU runtimes. Its stated focus is making machine-learning experiments easier to run, inspect, and assess—not claiming leaderboard performance. The project’s notebook list and current runnability have not been independently verified.
What is my_public_notebooks?
In a DEV Community article, Calvi presents my_public_notebooks as a public notebook lab built around a practical question: “How do I run it without guessing?” The notebooks are intended to run from beginning to end in Colab or on a local GPU runtime. The author says each should explain how to run it, what it demonstrates, and what it does not demonstrate.
That framing matters: the project is presented as an experimental and engineering resource, not as a collection of independently validated results. The available article text does not establish the repository’s current contents, license, commit history, or whether each notebook still runs.
Free tools Windows power users keep installed
One-click scans. No signup required.
What topics does the notebook collection cover?
Calvi’s article describes several distinct areas rather than one unified model or workflow:
#1 Best Overall
- Auditable Recursive Latent Reasoner (RLR) experiments: described as including independently computed ground truth, checksums, and held-out evaluation.
- Coconut and LFM2.5 production builds: described as continuous-thought approaches, with KV-cache optimization and structured-decision workloads.
- MiroFish, Graphiti, and Neo4j: local setup workflows in Colab, described as avoiding external API dependencies.
- DSPark: swarm and API benchmarks.
- HRM: product scenarios.
- Bonsai27: Colab workflows covering ngrok tunneling, CUDA fixes, and environment recipes.
These are the author’s descriptions of project areas, not confirmation that the experiments achieve particular outcomes or that the setups remain compatible with current software versions.
Why did the author open-source the notebooks?
Calvi gives three main motivations: making experiments more auditable, providing shared baselines for local-first and on-device language-model work, and documenting failure modes that papers or README files may leave out. These are reasons for sharing the work, not measured evidence that the notebooks have already produced those outcomes.
The guiding question “What does this prove?” is useful when reading any notebook in the collection. A reproducible run can show that a stated procedure works under its specified conditions; by itself, it does not establish broader model performance, generalize to other hardware, or validate an unstated claim.
How to contribute
The author invites contributions that improve reliability, comparisons, and practical documentation. Suggested work includes fixing broken cells, adding baselines, porting notebooks to CPU or Apple MPS, building production wrappers with timeouts, logging, and error handling, recording failure modes, and writing reproducible notebooks.
Before opening a pull request
- Open an issue first. Use it to describe the change and reduce the chance of duplicate work.
- Run the notebook in a fresh Colab runtime. This can reveal hidden dependencies on prior cell state or manually installed packages.
- Keep the change focused. A narrow fix or clearly scoped addition is easier to review and reproduce.
- Remove secrets and private data. Check notebook outputs, cells, and configuration before committing.
- Document new notebooks clearly. Explain how to run the notebook, what it is intended to demonstrate, and what it does not establish.
Contribution areas the author calls out
- Production wrappers, latency benchmarks, and on-device paths for LFM2.5 and Coconut.
- Alternative backends, memory persistence, and evaluation harnesses for Graphiti and Neo4j local pipelines.
- RLR comparisons with Mamba-2, GRU-RSSM, and transformer baselines under a common evaluation protocol.
- Colab environment recipes with pinned versions, install order, and T4 runtime caveats.
Those priorities are proposals in the author’s article; the available text does not provide a repository URL or confirm the present status of these tasks.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What is not established by the available project description?
The article describes intended scope and contribution practices, but it does not provide independent evidence for current notebook compatibility or experimental results. It also does not identify a repository link, license, or publication date in the captured text. Readers should therefore treat the listed components and workflows as author-described, and verify the project’s current repository details before relying on a notebook.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
