Torchrec adalah perpustakaan domain Pytorch yang dibangun untuk memberikan sparsity umum dan primitif paralelisme yang diperlukan untuk sistem rekomendasi skala besar (RECSYS). Torchrec memungkinkan pelatihan dan inferensi model dengan tabel embedding besar dicek di banyak GPU dan kekuatan banyak model recsys produksi di meta .
Torchrec telah digunakan untuk mempercepat kemajuan dalam sistem rekomendasi, beberapa contoh:
Untuk mulai belajar tentang Torchrec, lihat:
Lihat bagian Memulai dalam dokumentasi untuk cara yang disarankan untuk mengatur Torchrec.
Secara umum, tidak perlu membangun dari sumber . Untuk sebagian besar kasus penggunaan, ikuti bagian di atas untuk mengatur Torchrec. Namun, untuk membangun dari sumber dan untuk mendapatkan perubahan terbaru, lakukan hal berikut:
Instal Pytorch. Lihat Dokumentasi Pytorch.
CUDA 12.4
pip install torch --index-url https://download.pytorch.org/whl/nightly/cu124
CUDA 12.1
pip install torch --index-url https://download.pytorch.org/whl/nightly/cu121
CUDA 11.8
pip install torch --index-url https://download.pytorch.org/whl/nightly/cu118
CPU
pip install torch --index-url https://download.pytorch.org/whl/nightly/cpu
Klon Torchrec.
git clone --recursive https://github.com/pytorch/torchrec
cd torchrec
Instal FBGEMM.
CUDA 12.4
pip install fbgemm-gpu --index-url https://download.pytorch.org/whl/nightly/cu124
CUDA 12.1
pip install fbgemm-gpu --index-url https://download.pytorch.org/whl/nightly/cu121
CUDA 11.8
pip install fbgemm-gpu --index-url https://download.pytorch.org/whl/nightly/cu118
CPU
pip install fbgemm-gpu --index-url https://download.pytorch.org/whl/nightly/cpu
Pasang persyaratan lainnya.
pip install -r requirements.txt
Instal Torchrec.
python setup.py install develop
Uji instalasi (gunakan Torchx-Nightly untuk 3.11; untuk 3.12, Torchx saat ini tidak berfungsi).
GPU mode
torchx run -s local_cwd dist.ddp -j 1x2 --gpu 2 --script test_installation.py
CPU Mode
torchx run -s local_cwd dist.ddp -j 1x2 --script test_installation.py -- --cpu_only
Lihat Torchx untuk informasi lebih lanjut tentang meluncurkan pekerjaan terdistribusi dan jarak jauh.
Jika Anda ingin menjalankan contoh yang lebih kompleks, silakan lihat contoh Torchrec DLRM.
Lihat Kontribusi.MD untuk detail tentang berkontribusi pada Torchrec!
Jika Anda menggunakan Torchrec, silakan merujuk ke entri Bibtex untuk mengutip karya ini:
@inproceedings{10.1145/3523227.3547387,
author = {Ivchenko, Dmytro and Van Der Staay, Dennis and Taylor, Colin and Liu, Xing and Feng, Will and Kindi, Rahul and Sudarshan, Anirudh and Sefati, Shahin},
title = {TorchRec: a PyTorch Domain Library for Recommendation Systems},
year = {2022},
isbn = {9781450392785},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3523227.3547387},
doi = {10.1145/3523227.3547387},
abstract = {Recommendation Systems (RecSys) comprise a large footprint of production-deployed AI today. The neural network-based recommender systems differ from deep learning models in other domains in using high-cardinality categorical sparse features that require large embedding tables to be trained. In this talk we introduce TorchRec, a PyTorch domain library for Recommendation Systems. This new library provides common sparsity and parallelism primitives, enabling researchers to build state-of-the-art personalization models and deploy them in production. In this talk we cover the building blocks of the TorchRec library including modeling primitives such as embedding bags and jagged tensors, optimized recommender system kernels powered by FBGEMM, a flexible sharder that supports a veriety of strategies for partitioning embedding tables, a planner that automatically generates optimized and performant sharding plans, support for GPU inference and common modeling modules for building recommender system models. TorchRec library is currently used to train large-scale recommender models at Meta. We will present how TorchRec helped Meta’s recommender system platform to transition from CPU asynchronous training to accelerator-based full-sync training.},
booktitle = {Proceedings of the 16th ACM Conference on Recommender Systems},
pages = {482–483},
numpages = {2},
keywords = {information retrieval, recommender systems},
location = {Seattle, WA, USA},
series = {RecSys '22}
}
Torchrec berlisensi BSD, seperti yang ditemukan dalam file lisensi.