The errata list is a list of errors and their corrections that were found after the product was released.
The following errata were submitted by our customers and have not yet been approved or disproved by the author or editor. They solely represent the opinion of the customer.
Color Key: Serious technical mistake Minor technical mistake Language or formatting error Typo Question Note Update
Version |
Location |
Description |
Submitted by |
Date submitted |
Printed |
Page 5
3rd paragraph |
Hi
In the book chapter 2, you have the first example of Tensorflow for learning
You get data from:
from tensorflow.examples.tutorials.mnist import input_data
However this seem not really to be available anymore,
Do you have an update of your example code with
tf.keras.datasets.mnist.load_data()
|
Kell Michael Jensen |
Jan 10, 2022 |
Printed |
Page 16
2nd line of code |
27 from tensorflow.examples.tutorials.mnist import input_data
Error code below:
ModuleNotFoundError: No module named 'tensorflow.examples.tutorials'
Has the code been moved?
DO I need to add quotes
I am running it on Anaconda3 Jupitor.
Thanks for your help.
|
Jeffrey V Winston |
Dec 08, 2019 |
PDF, ePub |
Page 46
line 14 |
Earlier, `wb` is created as an empty list, but when appending the results of the current run, it is printed as:
> wb_.append(sess.run([w,b]))
And it should read:
> wb.append(sess.run([w,b]))
A similar error can be found on page 48
|
Andy Cyca |
May 01, 2019 |
Printed |
Page 74
2nd text paragraph |
A misplaced ( after hidden_layer_size
|
Aart Bik |
Nov 09, 2019 |
Printed |
Page 90
1st code block |
Hi,
Loving the book - I think I might have found a small error. It may have arisen due to the version of Tensorflow that I am using?
Let me know if you agree with my proposal.
python: 3.6.3
tensorflow==1.6.0
upon running the code as written this error is raised:
ValueError: Variable lstm/rnn/basic_lstm_cell/kernel already exists, disallowed. Did you mean to set reuse=True or reuse=tf.AUTO_REUSE in VarScope? Originally defined at:
This is caused due to the following:
If you use one variable twice (or more times), you should first time use with tf.variable_scope('scope_name', reuse=False): and the next time with tf.variable_scope('scope_name', reuse=True):.
Or you can use method tf.variable_scope.reuse_variables()
In this case the tf.variable_scope("lstm") is used twice without being explicitly reused.
To address I did the following:
```
with tf.variable_scope("lstm") as scope:
lstm_cell = tf.contrib.rnn.BasicLSTMCell(hidden_layer_size,
forget_bias=1.0)
scope.reuse_variables()
outputs, states = tf.nn.dynamic_rnn(lstm_cell, embed,
sequence_length = _seqlens,
dtype=tf.float32)
```
|
Tom Miller |
Apr 13, 2018 |
Printed |
Page 209
7th paragraph from the bottom |
When I changed from
opt = tf.train.DradientDecsentOptimizer(0.5)
to
opt = tf.train.AdamOptimizer(0.5)
an critical error appears.
saying,
“Attempting to use uninitialized value beta1_power_1”
Is there any way to properly initialize the Adam Optimizer in the class encapsulation style such as shown on the page 208-209?
|
Tomohisa Kumagai |
Dec 18, 2019 |