This is stars and forks stats for /NVIDIA-Merlin/dataloader repository. As of 29 Apr, 2024 this repository has 391 stars and 22 forks.
Merlin Dataloader The merlin-dataloader lets you quickly train recommender models for TensorFlow, PyTorch and JAX. It eliminates the biggest bottleneck in training recommender models, by providing GPU optimized dataloaders that read data directly into the GPU, and then do a 0-copy transfer to TensorFlow and PyTorch using dlpack. The benefits of the Merlin Dataloader include: Over 10x speedup over native framework dataloaders Handles larger than memory datasets Per-epoch shuffling Distributed training Installation Merlin-dataloader...
Merlin Dataloader The merlin-dataloader lets you quickly train recommender models for TensorFlow, PyTorch and JAX. It eliminates the biggest bottleneck in training recommender models, by providing GPU optimized dataloaders that read data directly into the GPU, and then do a 0-copy transfer to TensorFlow and PyTorch using dlpack. The benefits of the Merlin Dataloader include: Over 10x speedup over native framework dataloaders Handles larger than memory datasets Per-epoch shuffling Distributed training Installation Merlin-dataloader...
| repo | techs | stars | weekly | forks | weekly |
|---|---|---|---|---|---|
| comet-ml/kangas | Jupyter NotebookPythonJavaScript | 989 | 0 | 43 | 0 |
| estiens/world_cup_json | RubyHamlHTML | 930 | 0 | 109 | 0 |
| rustodon/rustodon | RustHTMLCSS | 831 | 0 | 44 | +1 |
| ohjelmistotuotanto-hy/ohjelmistotuotanto-hy.github.io | SCSSHTMLRuby | 5 | 0 | 85 | 0 |
| gloeyisk/universal-gms-doze | Shell | 1k | 0 | 25 | 0 |
| openshift-helm-charts/stage | SmartyMakefilePython | 2 | 0 | 9 | 0 |
| LeastAuthority/moonmath-manual | TeXShell | 252 | 0 | 46 | 0 |
| riscv-non-isa/riscv-rvm-csi | CPythonMakefile | 8 | 0 | 5 | 0 |
| bcgov/PSP | TSQLC#TypeScript | 7 | 0 | 24 | 0 |
| Oneplus/dotfiles | Vim ScriptPythonR | 0 | 0 | 2 | 0 |