Open tobeinged opened 1 year ago
Hi @tobeinged, of course you can read the chapters in any order, but here are my suggestions:
Chapter 4 is strictly about search algorithms and "classic" game-playing AIs. You can safely skip it for now if you're want to get straight to the deep learning. But the strongest game AIs today combine classic search algorithms with deep learning -- the final chapters show how this is done, so you may want to come back to chapter 4 before then
Chapter 5 is about the basics of neural networks. If you already have experience with deep learning and/or neural networks, you can safely skip it
Chapter 6 covers some important considerations about designing a neural network specifically for playing Go so I do recommend reading it before chapter 7
thank you~ @macfergus in the fact,i have go throught chapter1-8,and follow the code in the chapter 2,3;
but when i try to insert the code in the chapter 4 to impove my AI of GO,i find the code seems independent with the code before.It shows some excellent method in a lot games but dont realise then on the basis of the code before.
I can understand what the codes do,but i find it's hard for me to use them insert my codes before.I am afraid my agent may get 1000000000bugs because of my incorrect organizational form.
I download the resource from here and contrast them with my own codes,there are some new codes,and some new files.and i have to check one py file and then another, and i am not sure if i missed some codes.
i think every codelist in the books shows the path like dlgo/agent.py maybe better at the next version.
Anywhere,thankyou very much for your advice. I love this book and i love you too!
seems the code in chapter 4 just show some function of search so i dont have to update my code written at chapter 1 2 3 ? thanks for somebody's answer in the future, love you!