Transformers for 🇻🇦→🇬🇧 NMT

[Geremia]

Excellent explanation of Transformers, cited in Natural Language Processing in Action §9.2.2 as "a mind-expanding walk through the modern GPT architecture", "3Blue1Brown visualizations and explanations by Grant Sanderson":

(from full Neural Nets playlist)

[Geremia]

AquinasLatinEnglish / AquinasLatinEnglishModel uses Transformers and Byte-Pair Encoding.

original Transformers paper:

• Vaswani, Ashish, Noam Shazeer, Niki Parmar, et al. "Attention Is All You Need." arXiv:1706.03762. Preprint, arXiv, August 2, 2023 [1^st ed.: 2017].

original byte-pair encoding (BPE) paper:

• Sennrich, Rico, Barry Haddow, and Alexandra Birch. "Neural Machine Translation of Rare Words with Subword Units." arXiv:1508.07909. Preprint, arXiv, June 10, 2016.

[Geremia]

Latin 🇻🇦 → English 🇬🇧 Translator is now live. 😃

[Geremia]

original Transformers paper:

• Vaswani, Ashish, Noam Shazeer, Niki Parmar, et al. "Attention Is All You Need." arXiv:1706.03762. Preprint, arXiv, August 2, 2023 [1^st ed.: 2017].

Another good explanation of Transformers:

[Geremia]

3Blue1Brown

Another good video from 3Blue1Brown:

[Geremia]

But how do AI images and videos actually work? | Guest video by Welch Labs

cf. Comfy, the best open-source software for doing this, with excellent memory management; in fact, an 8 GB VRAM GPU can do some amazing generative AI videos.