Name: Build a DeepSeek Model (From Scratch)
Brand: Manning
SKU: 9781633434325
Availability: InStock

Build a DeepSeek Model (From Scratch)

PRE-ORDER NOW

skills. You will start by coding a DeepSeekAttention module, progress to building a fully functional MoE layer, and set up a high-efficiency training pipeline. By the end of the book, you will have a fully operational mini-DeepSeek that runs on your laptop, along with the skills to extend and optimize it for your own research or production applications.

SKU: 9781633434325 Category: Manning Brand: Manning

Description

Build a DeepSeek Model (From Scratch) uses intuitive visualizations, code walkthroughs, and a problem-solution narrative to transform complex concepts into practical skills. You will start by coding a DeepSeekAttention module, progress to building a fully functional MoE layer, and set up a high-efficiency training pipeline. By the end of the book, you will have a fully operational mini-DeepSeek that runs on your laptop, along with the skills to extend and optimize it for your own research or production applications.