Page 35, section 2.3.1

The code for the naive_relu() and naive_add() functions needs to use the full range of row and column indexes for iteration. The correct code is:

Page 79, listing 3.24
Page 90, listing 4.2

The code that demonstrates k-fold validation has an error in the call to cut(). The code should read as follows:

Page 145, figure 5.16

The PDF/eBook version of the book has an incorrect Figure 5.16: Training and validation metrics for fine-tuning. This figure was modified during editing and only the black and white version used in the print book was updated. The corrected figure is below:

Page 169, listing 6.4

In the word level one-hot encoding example, the invocation of the hash function should use index j rather than index 1:

Page 195, listing 6.33

The generator function has an issue with selection of rows (some observations at boundaries are excluded). The generator code should be updated to:

Page 223, listing 7.1

The code which creates the embedding layers should be modified as follows:

Page 225, listing 7.3

The code which defines embedded_posts should read:

Page 238, listing 7.9

The tensorboard example should be modified to remove the embeddings_freq parameter:

Page 262, listing 8.10

The calculation of the loss should be updated as follows:

Page 319, section A.4

The shell command for monitoring GPU utilization should be: