Wikipédia

PyTorch

PyTorch inggih punika pustaka machine learning ingkang adhedhasar pustaka Torch,[3][4][5] ingkang dipunginakaken kangge aplikasi kados pambelajaran jero,[6] ingkang asalipun dipunkembangaken dening Meta AI lan samenika dados bagean saking payung Linux Foundation.[7][8][9][10] Pustaka punika dipunakeni minangka salah satunggaling kalih pustaka machine learning ingkang paling populer sesarengan kaliyan TensorFlow, ingkang nawakaken piranti lunak sumber tinarbuka ingkang dipunrilis kanthi lisensi BSD ingkang dipunowahi. Sanajan interface Python langkung alus lan dados fokus utami pangembangan, PyTorch ugi gadhah interface C++.[11]

PyTorch
Perancang
  • Adam Paszke
  • Sam Gross
  • Soumith Chintala
  • Gregory Chanan
PangembangMeta AI
Rilis pisananSèptèmber 2016; 8 taun kapungkur (2016-09)[1]
Basa pamrograman
Sistem operasi
PlatformIA-32, x86-64, ARM64
JenisPakaryan machine learning and pambelajaran jero
LisènsiBSD-3[2]
Situs webpytorch.org

Meta (ingkang sakderengipun dipunsebat Facebook) ngoperasikaken PyTorch lan Convolutional Architecture for Fast Feature Embedding (Caffe2), nanging model-model ingkang dipundefinisikaken dening kalih framework punika boten saged dipunginakaken sesarengan. Proyek Open Neural Network Exchange (ONNX) dipundamel dening Meta lan Microsoft ing September 2017 kangge ngonversi model antawisipun framework. Caffe2 dipungabung kaliyan PyTorch ing pungkasaning Maret 2018.[12] Ing September 2022, Meta ngumumaken bilih PyTorch badhe dipunatur dening PyTorch Foundation, organisasi mandiri enggal ingkang dipundamel – anak perusahaan saking Linux Foundation.[13]

Ing PyTorch, jaringan saraf dipundamel ngagem inheritance saking modul torch.nn.Module. Modul punika nawakaken koleksi komprehensif saking blok bangunan kangge jaringan saraf, kalebet maneka warni lapisan lan fungsi aktivasi, ingkang ndadosaken konstruksi model ingkang kompleks. Sampunipun, proses komputasi input ing jaringan saraf menika ditindakaken ngeliwati urutan operasi ing fungsi forward().

PyTorch gadhah kathah keuntungan minangka framework pembelajaran mesin. Piyambakipun gadhah antarmuka ingkang gampil dipunginakaken lan intuitif ingkang saged nggampilaken proses ngembangaken model. PyTorch ugi nyengkuyung komputasi grafik dinamis, ingkang tegesipun grafik komputasi saged dipunewahi nalika running, ingkang migunani kangge kasus kados rekurens lan pemrosesan basa alami. Kejawi punika, PyTorch gadhah ekosistem ingkang ngrembaka saking alat lan perpustakaan pihak ketiga ingkang saged dipunginakaken kangge ngembangaken model ingkang langkung canggih. Sawetawis piranti lunak deep learning dipundamel ing sadhuwuring PyTorch, kalebet Tesla Autopilot,[14] Pyro-nipun Uber,[15] Transformers-ipun Hugging Face,[16] PyTorch Lightning,[17][18] lan Catalyst.[19]

Sitiran

besut
  1. ↑ Chintala, Soumith (1 September 2016). "PyTorch Alpha-1 release".
  2. ↑ Claburn, Thomas (12 September 2022). "PyTorch gets lit under The Linux Foundation". The Register.
  3. ↑ Yegulalp, Serdar (19 January 2017). "Facebook brings GPU-powered machine learning to Python". InfoWorld. Dibukak ing 11 December 2017.
  4. ↑ Lorica, Ben (3 August 2017). "Why AI and machine learning researchers are beginning to embrace PyTorch". O'Reilly Media. Dibukak ing 11 December 2017.
  5. ↑ Ketkar, Nikhil (2017). "Introduction to PyTorch". Deep Learning with Python (ing basa Inggris). Apress, Berkeley, CA. kc. 195–208. doi:10.1007/978-1-4842-2766-4_12. ISBN 9781484227657.
  6. ↑ Moez Ali (Jun 2023). "NLP with PyTorch: A Comprehensive Guide". datacamp.com (ing basa Inggris). Dibukak ing 2024-04-01.
  7. ↑ Patel, Mo (2017-12-07). "When two trends fuse: PyTorch and recommender systems". O'Reilly Media (ing basa Inggris). Dibukak ing 2017-12-18.
  8. ↑ Mannes, John. "Facebook and Microsoft collaborate to simplify conversions from PyTorch to Caffe2". TechCrunch (ing basa Inggris). Dibukak ing 2017-12-18. FAIR is accustomed to working with PyTorch – a deep learning framework optimized for achieving state of the art results in research, regardless of resource constraints. Unfortunately in the real world, most of us are limited by the computational capabilities of our smartphones and computers.
  9. ↑ Arakelyan, Sophia (2017-11-29). "Tech giants are using open source frameworks to dominate the AI community". VentureBeat (ing basa Inggris Amérika Sarékat). Dibukak ing 2017-12-18.
  10. ↑ "PyTorch strengthens its governance by joining the Linux Foundation". pytorch.org (ing basa Inggris). Dibukak ing 2022-09-13.
  11. ↑ "The C++ Frontend". PyTorch Master Documentation. Dibukak ing 2019-07-29.
  12. ↑ "Caffe2 Merges With PyTorch". 2018-04-02.
  13. ↑ Edwards, Benj (2022-09-12). "Meta spins off PyTorch Foundation to make AI framework vendor neutral". Ars Technica.
  14. ↑ Karpathy, Andrej. "PyTorch at Tesla - Andrej Karpathy, Tesla".
  15. ↑ "Uber AI Labs Open Sources Pyro, a Deep Probabilistic Programming Language". Uber Engineering Blog (ing basa Inggris Amérika Sarékat). 2017-11-03. Dibukak ing 2017-12-18.
  16. ↑ PYTORCH-TRANSFORMERS: PyTorch implementations of popular NLP Transformers, PyTorch Hub, 2019-12-01, dibukak ing 2019-12-01
  17. ↑ PYTORCH-Lightning: The lightweight PyTorch wrapper for ML researchers. Scale your models. Write less boilerplate, Lightning-Team, 2020-06-18, dibukak ing 2020-06-18
  18. ↑ "Ecosystem Tools". pytorch.org (ing basa Inggris). Dibukak ing 2020-06-18.
  19. ↑ "Ecosystem Tools". pytorch.org (ing basa Inggris). Dibukak ing 2020-04-04.
Dijupuk saka "https://jv.wikipedia.org/w/index.php?title=PyTorch&oldid=1676824"