Training LLM Models with Transformers

Alican Kiraz
4.7
12 ulasan
e-Buku
90
Halaman
Rating dan ulasan tidak disahkan  Ketahui Lebih Lanjut

Perihal e-buku ini

In our book, we explore in detail the working principles of artificial intelligence, LLM models, how these models can be trained, the training environment, how to prepare training data, and how to write the training code. We will learn in detail how to set up the training environment, use various Transformers libraries, train large LLM models through quantization (LoRA and QLoRA), create datasets in different formats, and grasp the mathematical foundations of artificial intelligence.

Rating dan ulasan

4.7
12 ulasan

Perihal pengarang

Hi, I'm Alican. I have been working in the field of cybersecurity since 2015. In my early years, I developed my skills in Application Security and Bug Hunting. Since 2018, I have been specializing in Blue Team and Purple Team roles. Because I am passionate about cybersecurity, I have participated in numerous training programs and certification courses, and I currently hold 23 different professional certifications. Additionally, I have been involved in various projects in robotics, embedded systems, and artificial intelligence. I support the community on YouTube and Spotify Podcast with more than 50 research and training videos. I also have 2 books and 2 Udemy courses.

Berikan rating untuk e-Buku ini

Beritahu kami pendapat anda.

Maklumat pembacaan

Telefon pintar dan tablet
Pasang apl Google Play Books untuk Android dan iPad/iPhone. Apl ini menyegerak secara automatik dengan akaun anda dan membenarkan anda membaca di dalam atau luar talian, walau di mana jua anda berada.
Komputer riba dan komputer
Anda boleh mendengar buku audio yang dibeli di Google Play menggunakan penyemak imbas web komputer anda.
eReader dan peranti lain
Untuk membaca pada peranti e-dakwat seperti Kobo eReaders, anda perlu memuat turun fail dan memindahkan fail itu ke peranti anda. Sila ikut arahan Pusat Bantuan yang terperinci untuk memindahkan fail ke e-Pembaca yang disokong.