Padding and sequence length in NLP - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to pad sequences to the same length using Keras.

NLP

from tensorflow.keras.preprocessing.sequence import pad_sequences

sequences = [[1, 2, 3], [4, 5], [6]]
padded = pad_sequences(sequences, maxlen=[1])
print(padded)

Drag options to blanks, or click blank then click option'

A3

B5

C2

D4

Attempts:

3 left

💡 Hint

Common Mistakes

Setting maxlen less than the longest sequence length, which truncates data.

Not using maxlen at all, causing inconsistent sequence lengths.

The longest sequence has length 3, so maxlen=3 pads all sequences to length 3.

2fill in blank

medium

Complete the code to pad sequences with zeros at the end (post-padding).

NLP

from tensorflow.keras.preprocessing.sequence import pad_sequences

sequences = [[7, 8], [9, 10, 11]]
padded = pad_sequences(sequences, padding=[1])
print(padded)

Drag options to blanks, or click blank then click option'

A'post'

B'pre'

C'middle'

D'end'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'pre' padding which adds zeros before sequences.

Using invalid padding values like 'middle' or 'end'.

Setting padding='post' adds zeros at the end of sequences.

3fill in blank

hard

Fix the error in the code to truncate sequences longer than maxlen from the beginning.

NLP

from tensorflow.keras.preprocessing.sequence import pad_sequences

sequences = [[1, 2, 3, 4], [5, 6, 7]]
padded = pad_sequences(sequences, maxlen=3, truncating=[1])
print(padded)

Drag options to blanks, or click blank then click option'

A'start'

B'post'

C'pre'

D'end'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'post' truncating which removes elements from the end.

Using invalid truncating values like 'start' or 'end'.

truncating='pre' removes elements from the start to fit maxlen.

4fill in blank

hard

Fill both blanks to create a padded sequence with max length 4, post-padding, and pre-truncating.

NLP

from tensorflow.keras.preprocessing.sequence import pad_sequences

sequences = [[10, 20, 30, 40, 50], [60, 70]]
padded = pad_sequences(sequences, maxlen=[1], padding=[2], truncating='pre')
print(padded)

Drag options to blanks, or click blank then click option'

A4

B'post'

C'pre'

D5

Attempts:

3 left

💡 Hint

Common Mistakes

Using maxlen larger than needed, causing no truncation.

Using 'pre' padding which adds zeros before sequences.

maxlen=4 limits length to 4; padding='post' adds zeros after sequences.

5fill in blank

hard

Fill all three blanks to create a padded sequence with max length 3, pre-padding, and post-truncating.

NLP

from tensorflow.keras.preprocessing.sequence import pad_sequences

sequences = [[100, 200, 300, 400], [500]]
padded = pad_sequences(sequences, maxlen=[1], padding=[2], truncating=[3])
print(padded)

Drag options to blanks, or click blank then click option'

A3

B'pre'

C'post'

D'start'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'post' padding which adds zeros after sequences.

Using 'pre' truncating which removes from the start.

maxlen=3 limits length; padding='pre' adds zeros before; truncating='post' removes from end.