Custom pipeline components in NLP - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to define a custom pipeline component function that takes a doc and returns it.

NLP

def custom_component(doc):
    # Process the doc here
    return [1]

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

Atext

Btokens

Cdoc

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Returning None instead of the doc object.

Returning the text string instead of the doc object.

The custom pipeline component must return the processed doc object to continue the pipeline.

2fill in blank

medium

Complete the code to add the custom component to the pipeline named 'custom_component'.

NLP

nlp.add_pipe([1], name='custom_component', last=True)

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

Apipeline

Bprocess_doc

Ccomponent

Dcustom_component

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a string instead of the function name.

Using a wrong function name that is not defined.

You add the custom component function by passing its name to add_pipe.

3fill in blank

hard

Fix the error in the custom component that tries to add an attribute 'is_positive' to each token.

NLP

def custom_component(doc):
    for token in doc:
        token.[1] = token.text.startswith('good')
    return doc

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

Apos

B_.is_positive

Cis_positive

Dset_attr

Attempts:

3 left

💡 Hint

Common Mistakes

Trying to set token.is_positive directly without registering the attribute.

Using a wrong attribute name or syntax.

Custom token attributes must be set using the underscore accessor '_.attribute_name'.

4fill in blank

hard

Fill both blanks to register a custom token attribute 'is_positive' with default False.

NLP

from spacy.tokens import Token
Token.set_extension('[1]', default=[2])

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

Ais_positive

BTrue

CFalse

Dsentiment

Attempts:

3 left

💡 Hint

Common Mistakes

Using True as default when you want False.

Using a different attribute name than used in the component.

You register the attribute name 'is_positive' with default value False.

5fill in blank

hard

Fill all three blanks to create a custom pipeline component that sets 'is_positive' on tokens and returns the doc.

NLP

def [1](doc):
    for token in doc:
        token._.[2] = token.text.lower() in [3]
    return doc

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

Acustom_component

Bis_positive

C{'good', 'great', 'happy'}

Dpositive_tokens

Attempts:

3 left

💡 Hint

Common Mistakes

Using a different function name than registered.

Not using a set for the positive words.

Forgetting to return the doc.

The function name is 'custom_component', attribute is 'is_positive', and the set of positive words is {'good', 'great', 'happy'}.