Interactive deep learning book with multi-framework code, math, and discussions. Adopted at 500 universities from 70 countries including Stanford, MIT, Harvard, and Cambridge.
First, I would like an easier way to download interactive version of the book - why manuals and books these days cannot be downloaded just by one click? :P
Second, when I copy the code from the book, it will not keep indents - it's hard to copy from PDF. It should be structured differently to keep the spaces and empty lines of Python code.
Third, for some reason, if I download TensorFlow PDF, it still contains PyTorch code.
Fourth, in d2l pip download, there are not all @saved functions. I tried this (for example from page 143):
from d2l import torch as d2l
d2l.train_ch3
Traceback (most recent call last):
File "", line 1, in
AttributeError: module 'd2l.torch' has no attribute 'train_ch3'. Did you mean: 'train_ch13'?
Fifth, as I am often in place without internet connection, it would help me a lot if there was a way to pre-download all datasets used in a book, kind of like d2l, for example with one command, and then use those cached datasets. Maybe two commands "cache_datasets" and "use_cached_datasets".
I think I haven't ran into any more troubles with this book.
Hello!
I take them all together in one post.
First, I would like an easier way to download interactive version of the book - why manuals and books these days cannot be downloaded just by one click? :P
Second, when I copy the code from the book, it will not keep indents - it's hard to copy from PDF. It should be structured differently to keep the spaces and empty lines of Python code.
Third, for some reason, if I download TensorFlow PDF, it still contains PyTorch code.
Fourth, in d2l pip download, there are not all @saved functions. I tried this (for example from page 143):
Fifth, as I am often in place without internet connection, it would help me a lot if there was a way to pre-download all datasets used in a book, kind of like d2l, for example with one command, and then use those cached datasets. Maybe two commands "cache_datasets" and "use_cached_datasets".
I think I haven't ran into any more troubles with this book.