The Author Online Book Forums are Moving

The Author Online Book Forums will soon redirect to Manning's liveBook and liveVideo. All book forum content will migrate to liveBook's discussion forum and all video forum content will migrate to liveVideo. Log in to liveBook or liveVideo with your Manning credentials to join the discussion!

Thank you for your engagement in the AoF over the years! We look forward to offering you a more enhanced forum experience.

Susan Harkins (424) [Avatar] Offline
#101
An updated errata list for Deep Learning with Python is available at https://manning-content.s3.amazonaws.com/download/9/972b1ae-dfc3-47c1-80d0-afa55c4b9415/Chollet_DeepLearninginPython_err3.html. The pbook has been updated and corrections to the mobile and livebook versions are pending. Thank you for participating in the collection process. Your contributions are a great help to us and other readers.

Regards,
Susan Harkins
Errata Editor
277082 (1) [Avatar] Offline
#102
page 40, first sentence (cont'd from previous page):

"[...] repeated 10 times alongside a new axis is a helpful mental model."

to

"[...] repeated 32 times alongside a new axis is a helpful mental model."

On another note,
@Susan Harkins: Could you maybe convince Manning to make Chapter 2 and 3 freely available for everyone in the Livebook section just like they did with Chapter 1 ? (reason being, all 3 Chapters are freely available as pdf in the Sample Chapters on https://www.manning.com/books/deep-learning-with-python anyway).

Specifically:
https://livebook.manning.com/#!/book/deep-learning-with-python/chapter-2/
https://livebook.manning.com/#!/book/deep-learning-with-python/chapter-3/
Susan Harkins (424) [Avatar] Offline
#103
I'll pass along your request -- thanks for letting me know.

Regards,
Susan Harkins
Errata Editor
Susan Harkins (424) [Avatar] Offline
#104
We still offer free chapters via PDF, and they can be found in the "book resources" dropdown menu, but we replaced free chapters on liveBook with the time limited free preview. Now instead of specified free chapters, readers can preview the entire book for 5 minutes per day.

Regards,

Susan Harkins
Errata Editor
Haesun Park (24) [Avatar] Offline
#105
( p248 ) In the last code block of 7.1.6,

right_input = xception_base(right_input)
should be
right_features = xception_base(right_input)


[left_features, right_input], axis=-1)
should be
[left_features, right_features], axis=-1)
636740 (1) [Avatar] Offline
#106
Tiny one, the chain rule on p51 is described as (second line of second paragraph of 2.4.4):

f(g(x)) = f'(g(x)) * g'(x)


I think this is missing the differentiation on the left:

f(g(x))' = f'(g(x)) * g'(x)
Kyungwon Chun (24) [Avatar] Offline
#107
Loc 618
This form of data analy-sis
->
This form of data analysis
Kyungwon Chun (24) [Avatar] Offline
#108
Loc 769
natural--language
->
natural-language
Kyungwon Chun (24) [Avatar] Offline
#109
Loc 1035
(-samples, frames, height, width, color_depth)
->
(samples, frames, height, width, color_depth)
Kyungwon Chun (24) [Avatar] Offline
#110
Loc 6983
sudo apt-get install python-numpy python-scipy python- matplotlib
python-yaml
->
sudo apt-get install python-numpy python-scipy python-matplotlib python-yaml
Kyungwon Chun (24) [Avatar] Offline
#111
Loc 7016
$ sudo dpkg -i dpkg -i libcudnn6*.deb

->
$ sudo dpkg -i libcudnn6*.deb
Kyungwon Chun (24) [Avatar] Offline
#112
Loc 7046
watch -n 5 NVIDIA-smi -a --display=utilization

->
watch -n 5 nvidia-smi -a --display=utilization
Kyungwon Chun (24) [Avatar] Offline
#113
Loc 1719
plt.ylabel('Loss')

->
plt.ylabel('Accuracy')
Kyungwon Chun (24) [Avatar] Offline
#114
Loc 1897
sparse_categorical_-crossentropy
->
sparse_categorical_crossentropy
Kyungwon Chun (24) [Avatar] Offline
#115
At Listing 5.18

model.add(layers.Dense(256, activation='relu', input_shape=4 * 4 * 512))

->
model.add(layers.Dense(256, activation='relu', input_shape=(4 * 4 * 512,)))
Kyungwon Chun (24) [Avatar] Offline
#116
At Listing 5.25,

The first comment arrow should point to the following line
img_tensor = np.expand_dims(img_tensor, axis=0)


The following line should be fixed,
<1> Its shape is (1, 150,1 50, 3)

->
# Its shape is (1, 150,1 50, 3)


Also, the second comment arrow should point to the following line
# Its shape is (1, 150,1 50, 3)
Kyungwon Chun (24) [Avatar] Offline
#117
At Loc 3425
You brain has learned
->
Your brain has learned
Kyungwon Chun (24) [Avatar] Offline
#118
Loc 3678

natural-l-anguage
->
natural-language
Kyungwon Chun (24) [Avatar] Offline
#119
Loc 3812,
legal--document-classification model
->
legal-document-classification model
Kyungwon Chun (24) [Avatar] Offline
#120
At Listing 6.2, the following two lines of code should be fixed.
token_index = dict(zip(range(1, len(characters) + 1), characters))

should be
token_index = dict(zip(characters, range(1, len(characters) + 1)))

Also,
results = np.zeros(shape=(len(samples), max_length, max(token_index.keys()) + 1))

should be
results = np.zeros(shape=(len(samples), max_length, max(token_index.values()) + 1))
Kyungwon Chun (24) [Avatar] Offline
#121
At loc 3972,
a task--specific
->
a task-specific
Kyungwon Chun (24) [Avatar] Offline
#122
At loc 3999,
natural--language-processing
->
natural-language-processing
Kyungwon Chun (24) [Avatar] Offline
#123
At loc 4228,
natural--language-processing
->
natural-language-processing
Kyungwon Chun (24) [Avatar] Offline
#124
At loc 5090, the title of section 7.2.
callba- acks
->
callbacks
Kyungwon Chun (24) [Avatar] Offline
#125
At loc 5146,
loss plateau is is
->
loss plateau is