Why spaCy is production-grade NLP - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load a spaCy model for English.

NLP

import spacy
nlp = spacy.load('[1]')

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

Aenglish_model

Bspacy_en

Cen_nlp

Den_core_web_sm

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect or made-up model names.

Forgetting to install the model before loading.

The correct model name to load the small English model in spaCy is en_core_web_sm.

2fill in blank

medium

Complete the code to process text and get tokens using spaCy.

NLP

doc = nlp('[1]')
tokens = [token.text for token in doc]

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

Atoken.text

Bnlp

CThis is a test.

Ddoc

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the nlp object itself instead of text.

Passing token attributes instead of raw text.

You need to pass a string of text to the nlp object to process it and get tokens.

3fill in blank

hard

Fix the error in the code to get named entities from a spaCy doc.

NLP

for ent in doc.[1]:
    print(ent.text, ent.label_)

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

Aentities

Bents

Ctokens

Dnamed_entities

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'entities' or 'named_entities' which do not exist.

Trying to iterate over 'tokens' for entities.

The correct attribute to access named entities in a spaCy doc is ents.

4fill in blank

hard

Fill both blanks to create a dictionary of token texts and their parts of speech.

NLP

pos_dict = {token.[1]: token.[2] for token in doc}

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

Atext

Bpos_

Clemma_

Dtag_

Attempts:

3 left

💡 Hint

Common Mistakes

Using lemma_ instead of text for keys.

Using tag_ instead of pos_ for parts of speech.

Use text for token text and pos_ for part of speech tags.

5fill in blank

hard

Fill all three blanks to filter tokens that are alphabetic and lowercase their text.

NLP

filtered = [token.[1].lower() for token in doc if token.[2] and not token.[3]]

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

Atext

Bis_alpha

Cis_stop

Dis_punct

Attempts:

3 left

💡 Hint

Common Mistakes

Using is_stop instead of is_punct to exclude tokens.

Not converting text to lowercase.

We use text to get token text, is_alpha to check if token is alphabetic, and exclude punctuation with is_punct.

Practice

1. Why is spaCy considered production-grade NLP?

easy

A. Because it is fast, accurate, and ready for real-world use

B. Because it only supports English language

C. Because it requires manual model training for every task

D. Because it is mainly for academic research, not applications

Solution

1. Step 1: Understand spaCy's design goals

   spaCy is built to be fast and accurate for practical NLP tasks.

2. Step 2: Identify production features

   It offers ready-to-use models and clear structure for building apps.

3. Final Answer:

   Because it is fast, accurate, and ready for real-world use -> Option A

4. Quick Check:

   Production-grade = Fast + Accurate + Ready [OK]

Hint: Look for speed, accuracy, and real-world readiness [OK]

Common Mistakes:

• Thinking spaCy supports only English
• Assuming manual training is always needed
• Confusing research tools with production tools

2. Which of the following is the correct way to load a spaCy English model in Python?

easy

A. import spacy; nlp = spacy.load('en_core_web_sm')

B. import spacy; nlp = spacy.load_model('english')

C. from spacy import load; nlp = load('en')

D. import spacy; nlp = spacy.load('english_model')

Solution

1. Step 1: Recall spaCy model loading syntax

   The correct function is spacy.load() with the model name string.

2. Step 2: Identify the official English model name

   The standard small English model is 'en_core_web_sm'.

3. Final Answer:

   import spacy; nlp = spacy.load('en_core_web_sm') -> Option A

4. Quick Check:

   Use spacy.load('en_core_web_sm') [OK]

Hint: Use spacy.load with exact model name string [OK]

Common Mistakes:

• Using incorrect function names like load_model
• Using wrong model names like 'english'
• Confusing import statements

3. What will be the output of this code snippet?

import spacy
nlp = spacy.load('en_core_web_sm')
doc = nlp('Apple is looking at buying a startup in the UK.')
print([(ent.text, ent.label_) for ent in doc.ents])

medium

A. [('Apple', 'PERSON'), ('UK', 'COUNTRY')]

B. []

C. [('Apple', 'ORG'), ('startup', 'ORG')]

D. [('Apple', 'ORG'), ('UK', 'GPE')]

Solution

1. Step 1: Understand spaCy named entity recognition

   spaCy identifies 'Apple' as an organization and 'UK' as a geopolitical entity.

2. Step 2: Check the entities extracted from the sentence

   Entities are [('Apple', 'ORG'), ('UK', 'GPE')].

3. Final Answer:

   [('Apple', 'ORG'), ('UK', 'GPE')] -> Option D

4. Quick Check:

   Entities = [('Apple', 'ORG'), ('UK', 'GPE')] [OK]

Hint: Look for common named entities like ORG and GPE [OK]

Common Mistakes:

• Confusing PERSON with ORG for 'Apple'
• Expecting 'startup' as an entity
• Assuming no entities detected

4. Identify the error in this spaCy code snippet:

import spacy
nlp = spacy.load('en_core_web_sm')
doc = nlp('Hello world')
for token in doc.tokens:
    print(token.text)

medium

A. The model name 'en_core_web_sm' is incorrect

B. The attribute 'tokens' does not exist on the doc object

C. Missing parentheses in print statement

D. The 'nlp' object is not callable

Solution

1. Step 1: Check spaCy Doc object attributes

   The Doc object uses 'doc' itself as iterable, not 'doc.tokens'.

2. Step 2: Identify correct iteration method

   Use 'for token in doc:' instead of 'doc.tokens'.

3. Final Answer:

   The attribute 'tokens' does not exist on the doc object -> Option B

4. Quick Check:

   Doc.tokens attribute error [OK]

Hint: Iterate directly over doc, not doc.tokens [OK]

Common Mistakes:

• Using doc.tokens instead of doc
• Incorrect model name assumption
• Forgetting print parentheses

5. You want to build a fast app that extracts entities from multiple languages using spaCy. Which feature makes spaCy production-grade for this task?

hard

A. spaCy only supports English and requires external tools for other languages

B. spaCy requires training a new model from scratch for each language

C. spaCy provides pre-trained models for many languages with optimized pipelines

D. spaCy uses slow but highly accurate models unsuitable for real-time apps

Solution

1. Step 1: Understand spaCy's multilingual support

   spaCy offers pre-trained models for many languages ready to use.

2. Step 2: Recognize production features for speed and accuracy

   These models have optimized pipelines for fast processing in apps.

3. Final Answer:

   spaCy provides pre-trained models for many languages with optimized pipelines -> Option C

4. Quick Check:

   Pre-trained multilingual models = production-ready [OK]

Hint: Choose pre-trained multilingual models for speed [OK]

Common Mistakes:

• Thinking all models must be trained from scratch
• Assuming spaCy supports only English
• Believing spaCy models are too slow for apps