# keskarnitish/large-batch-training

<https://repo.masters-masters.com/en/repos/keskarnitish/large-batch-training>

[Home](https://repo.masters-masters.com/en) › [keskarnitish](https://repo.masters-masters.com/en/owners/keskarnitish) › [large-batch-training](https://repo.masters-masters.com/en/repos/keskarnitish/large-batch-training)

keskarnitish/large-batch-training has 147 stars on GitHub. Its RepoMasters card: Prototype tier. Code to reproduce some of the figures in the paper "On…

| Field | Value |
| --- | --- |
| Owner | [keskarnitish](https://repo.masters-masters.com/en/owners/keskarnitish) |
| Stars | 147 |
| Forks | 23 |
| Language | [Python](https://repo.masters-masters.com/en/languages/python) |
| License | MIT |
| Created | 2016 |
| Topics | deep-learning, minima, batch-training, theano, keras |
| Tier | [Prototype (100 to 249 ★)](https://repo.masters-masters.com/en/ranks/rush) |
| View on GitHub | [keskarnitish/large-batch-training](https://github.com/keskarnitish/large-batch-training) |

## Repository

Code to reproduce some of the figures in the paper "On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima"

The tier for repos with 100 to 249 ★. Published odds per card pulled: 70%. Rarity: Common.

## Nearby in Python

- [THUROI0787/absent-author · ★ 147 · Python · Prototype](https://repo.masters-masters.com/en/repos/THUROI0787/absent-author)
- [galilai-group/prh-proof · ★ 147 · Python · Prototype](https://repo.masters-masters.com/en/repos/galilai-group/prh-proof)
- [AbdullahAlhariri/penmentor · ★ 147 · Python · Prototype](https://repo.masters-masters.com/en/repos/AbdullahAlhariri/penmentor)
- [pythonexcels/examples · ★ 147 · Python · Prototype](https://repo.masters-masters.com/en/repos/pythonexcels/examples)
- [hudora/pyShipping · ★ 147 · Python · Prototype](https://repo.masters-masters.com/en/repos/hudora/pyShipping)
- [indygreg/collectd-carbon · ★ 147 · Python · Prototype](https://repo.masters-masters.com/en/repos/indygreg/collectd-carbon)

---
Data: GitHub · Illustration drawn by RepoMasters from the repo’s ID and its language color.