google-research/tuning_playbook
View on GitHub ↗ ★ 30,417
A playbook for systematically maximizing the performance of deep learning models.
12 mentions on Hacker News by 12 people: 6 stories, 6 comments · first 2023-01-19 · latest 2025-08-22
20082026
-
You might find this helpful for prioritizing which knobs to turn first https://github.com/google-research/tuning_playbook
comment by lamename · 2025-03-26 · in Optimizing ML training with metagradient descent -
Maybe this will help: https://github.com/google-research/tuning_playbook
comment by incognito124 · 2025-03-21 · in Pen and Paper Exercises in Machine Learning (2022) -
I think you might want to read the Tuning Playbook if you haven't already. https://github.com/google-research/tuning_playbook
comment by yorwba · 2024-04-18 · in Ask HN: Are deep learning theorist employable in the industry -
…numbers. Despite games, another use case is for hyperparameter search for neural networks. https://github.com/google-research/tuning_playbook?tab=readm...
comment by albertzeyer · 2024-01-08 · in When random numbers are too random: Low discrepancy sequences (2017) -
…with Hyperband [3] for DL. [1] https://wires.onlinelibrary.wiley.com/doi/full/10.1002/widm.... [2] https://github.com/google-research/tuning_playbook [3] https://jmlr.csail.mit.edu/papers/v18/16-558.html
comment by blackbear_ · 2023-06-06 · in Hyperparameter Optimization for LLMs via Scaling Laws -
…the same final performance should be attainable using any batch size (see Shallue et al. 2018). https://github.com/google-research/tuning_playbook#choosing-... The ideal case is full-batch with tuneable regularisation, just the hardware…
comment by DougBTX · 2023-04-14 · in Beyond automatic differentiation