Deep Learning is evolving faster than ever. Every week we see new techniques: LoRA QLoRA RAG Foundation Models Mamba Diffusion Models But one question is still difficult: Which technique should I choose for my problem? Most resources explai...
Deep Learning is evolving faster than ever.
Every week we see new techniques:
LoRA
QLoRA
RAG
Foundation Models
Mamba
Diffusion Models
But one question is still difficult:
Which technique should I choose for my problem?
Most resources explain how something works.
Very few explain:
Why was it created?
When should I use it?
When should I avoid it?
What are the trade-offs?
That is the problem I wanted to solve.
Introducing: Deep Learning Training Playbook
A practical guide to help engineers make better AI decisions.
The first part covers:
✅ Transfer Learning
✅ Foundation Models
✅ Full Fine-Tuning
✅ LoRA
✅ QLoRA
✅ Continued Pretraining
✅ Knowledge Distillation
The goal:
Don't memorize AI techniques. Learn when and why to use them.
Open Source
GitHub:
https://github.com/wildoctopus/deep-learning-training-playbook
More chapters coming:
Model Architectures
Generative AI
Training Optimization
Agentic AI
Production AI Systems
Built by WildOctopus 🐙
Turning research papers into practical engineering playbooks.