Get started

spaCy 101: Everything you need to know

The most important concepts, explained in simple terms

Whether you’re new to spaCy, or just want to brush up on some NLP basics and implementation details – this page should have you covered. Each section will explain one of spaCy’s features in simple terms and with examples or illustrations. Some sections will also reappear across the usage guides as a quick introduction.

What’s spaCy?

spaCy is a free, open-source library for advanced Natural Language Processing (NLP) in Python.

If you’re working with a lot of text, you’ll eventually want to know more about it. For example, what’s it about? What do the words mean in context? Who is doing what to whom? What companies and products are mentioned? Which texts are similar to each other?

spaCy is designed specifically for production use and helps you build applications that process and “understand” large volumes of text. It can be used to build information extraction or natural language understanding systems, or to pre-process text for deep learning.

What spaCy isn’t

  • spaCy is not a platform or “an API”. Unlike a platform, spaCy does not provide a software as a service, or a web application. It’s an open-source library designed to help you build NLP applications, not a consumable service.

  • spaCy is not an out-of-the-box chat bot engine. While spaCy can be used to power conversational applications, it’s not designed specifically for chat bots, and only provides the underlying text processing capabilities.

  • spaCy is not research software. It’s built on the latest research, but it’s designed to get things done. This leads to fairly different design decisions than NLTK or CoreNLP, which were created as platforms for teaching and research. The main difference is that spaCy is integrated and opinionated. spaCy tries to avoid asking the user to choose between multiple algorithms that deliver equivalent functionality. Keeping the menu small lets spaCy deliver generally better performance and developer experience.

  • spaCy is not a company. It’s an open-source library. Our company publishing spaCy and other software is called Explosion AI.


In the documentation, you’ll come across mentions of spaCy’s features and capabilities. Some of them refer to linguistic concepts, while others are related to more general machine learning functionality.

TokenizationSegmenting text into words, punctuations marks etc.
Part-of-speech (POS) TaggingAssigning word types to tokens, like verb or noun.
Dependency ParsingAssigning syntactic dependency labels, describing the relations between individual tokens, like subject or object.
LemmatizationAssigning the base forms of words. For example, the lemma of “was” is “be”, and the lemma of “rats” is “rat”.
Sentence Boundary Detection (SBD)Finding and segmenting individual sentences.
Named Entity Recognition (NER)Labelling named “real-world” objects, like persons, companies or locations.
Entity Linking (EL)Disambiguating textual entities to unique identifiers in a Knowledge Base.
SimilarityComparing words, text spans and documents and how similar they are to each other.
Text ClassificationAssigning categories or labels to a whole document, or parts of a document.
Rule-based MatchingFinding sequences of tokens based on their texts and linguistic annotations, similar to regular expressions.
TrainingUpdating and improving a statistical model’s predictions.
SerializationSaving objects to files or byte strings.

Statistical models

While some of spaCy’s features work independently, others require statistical models to be loaded, which enable spaCy to predict linguistic annotations – for example, whether a word is a verb or a noun. spaCy currently offers statistical models for a variety of languages, which can be installed as individual Python modules. Models can differ in size, speed, memory usage, accuracy and the data they include. The model you choose always depends on your use case and the texts you’re working with. For a general-purpose use case, the small, default models are always a good start. They typically include the following components:

  • Binary weights for the part-of-speech tagger, dependency parser and named entity recognizer to predict those annotations in context.
  • Lexical entries in the vocabulary, i.e. words and their context-independent attributes like the shape or spelling.
  • Data files like lemmatization rules and lookup tables.
  • Word vectors, i.e. multi-dimensional meaning representations of words that let you determine how similar they are to each other.
  • Configuration options, like the language and processing pipeline settings, to put spaCy in the correct state when you load in the model.

Linguistic annotations

spaCy provides a variety of linguistic annotations to give you insights into a text’s grammatical structure. This includes the word types, like the parts of speech, and how the words are related to each other. For example, if you’re analyzing text, it makes a huge difference whether a noun is the subject of a sentence, or the object – or whether “google” is used as a verb, or refers to the website or company in a specific context.

Once you’ve downloaded and installed a model, you can load it via spacy.load(). This will return a Language object containing all components and data needed to process text. We usually call it nlp. Calling the nlp object on a string of text will return a processed Doc:

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("Apple is looking at buying U.K. startup for $1 billion")
for token in doc:
    print(token.text, token.pos_, token.dep_)

Even though a Doc is processed – e.g. split into individual words and annotated – it still holds all information of the original text, like whitespace characters. You can always get the offset of a token into the original string, or reconstruct the original by joining the tokens and their trailing whitespace. This way, you’ll never lose any information when processing text with spaCy.


During processing, spaCy first tokenizes the text, i.e. segments it into words, punctuation and so on. This is done by applying rules specific to each language. For example, punctuation at the end of a sentence should be split off – whereas “U.K.” should remain one token. Each Doc consists of individual tokens, and we can iterate over them:

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("Apple is looking at buying U.K. startup for $1 billion")
for token in doc:

First, the raw text is split on whitespace characters, similar to text.split(' '). Then, the tokenizer processes the text from left to right. On each substring, it performs two checks:

  1. Does the substring match a tokenizer exception rule? For example, “don’t” does not contain whitespace, but should be split into two tokens, “do” and “n’t”, while “U.K.” should always remain one token.

  2. Can a prefix, suffix or infix be split off? For example punctuation like commas, periods, hyphens or quotes.

If there’s a match, the rule is applied and the tokenizer continues its loop, starting with the newly split substrings. This way, spaCy can split complex, nested tokens like combinations of abbreviations and multiple punctuation marks.

Example of the tokenization process

While punctuation rules are usually pretty general, tokenizer exceptions strongly depend on the specifics of the individual language. This is why each available language has its own subclass like English or German, that loads in lists of hard-coded data and exception rules.

Part-of-speech tags and dependencies Needs model

After tokenization, spaCy can parse and tag a given Doc. This is where the statistical model comes in, which enables spaCy to make a prediction of which tag or label most likely applies in this context. A model consists of binary data and is produced by showing a system enough examples for it to make predictions that generalize across the language – for example, a word following “the” in English is most likely a noun.

Linguistic annotations are available as Token attributes. Like many NLP libraries, spaCy encodes all strings to hash values to reduce memory usage and improve efficiency. So to get the readable string representation of an attribute, we need to add an underscore _ to its name:

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("Apple is looking at buying U.K. startup for $1 billion")

for token in doc:
    print(token.text, token.lemma_, token.pos_, token.tag_, token.dep_,
            token.shape_, token.is_alpha, token.is_stop)

Using spaCy’s built-in displaCy visualizer, here’s what our example sentence and its dependencies look like:

Named Entities Needs model

A named entity is a “real-world object” that’s assigned a name – for example, a person, a country, a product or a book title. spaCy can recognize various types of named entities in a document, by asking the model for a prediction. Because models are statistical and strongly depend on the examples they were trained on, this doesn’t always work perfectly and might need some tuning later, depending on your use case.

Named entities are available as the ents property of a Doc:

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("Apple is looking at buying U.K. startup for $1 billion")

for ent in doc.ents:
    print(ent.text, ent.start_char, ent.end_char, ent.label_)
Apple05ORGCompanies, agencies, institutions.
U.K.2731GPEGeopolitical entity, i.e. countries, cities, states.
$1 billion4454MONEYMonetary values, including unit.

Using spaCy’s built-in displaCy visualizer, here’s what our example sentence and its named entities look like:

Word vectors and similarity Needs model

Similarity is determined by comparing word vectors or “word embeddings”, multi-dimensional meaning representations of a word. Word vectors can be generated using an algorithm like word2vec and usually look like this:


array([2.02280000e-01, -7.66180009e-02, 3.70319992e-01, 3.28450017e-02, -4.19569999e-01, 7.20689967e-02, -3.74760002e-01, 5.74599989e-02, -1.24009997e-02, 5.29489994e-01, -5.23800015e-01, -1.97710007e-01, -3.41470003e-01, 5.33169985e-01, -2.53309999e-02, 1.73800007e-01, 1.67720005e-01, 8.39839995e-01, 5.51070012e-02, 1.05470002e-01, 3.78719985e-01, 2.42750004e-01, 1.47449998e-02, 5.59509993e-01, 1.25210002e-01, -6.75960004e-01, 3.58420014e-01, # ... and so on ... 3.66849989e-01, 2.52470002e-03, -6.40089989e-01, -2.97650009e-01, 7.89430022e-01, 3.31680000e-01, -1.19659996e+00, -4.71559986e-02, 5.31750023e-01], dtype=float32)

Models that come with built-in word vectors make them available as the Token.vector attribute. Doc.vector and Span.vector will default to an average of their token vectors. You can also check if a token has a vector assigned, and get the L2 norm, which can be used to normalize vectors.

import spacy

nlp = spacy.load("en_core_web_md")
tokens = nlp("dog cat banana afskfsd")

for token in tokens:
    print(token.text, token.has_vector, token.vector_norm, token.is_oov)

The words “dog”, “cat” and “banana” are all pretty common in English, so they’re part of the model’s vocabulary, and come with a vector. The word “afskfsd” on the other hand is a lot less common and out-of-vocabulary – so its vector representation consists of 300 dimensions of 0, which means it’s practically nonexistent. If your application will benefit from a large vocabulary with more vectors, you should consider using one of the larger models or loading in a full vector package, for example, en_vectors_web_lg, which includes over 1 million unique vectors.

spaCy is able to compare two objects, and make a prediction of how similar they are. Predicting similarity is useful for building recommendation systems or flagging duplicates. For example, you can suggest a user content that’s similar to what they’re currently looking at, or label a support ticket as a duplicate if it’s very similar to an already existing one.

Each Doc, Span and Token comes with a .similarity() method that lets you compare it with another object, and determine the similarity. Of course similarity is always subjective – whether “dog” and “cat” are similar really depends on how you’re looking at it. spaCy’s similarity model usually assumes a pretty general-purpose definition of similarity.

import spacy

nlp = spacy.load("en_core_web_md")  # make sure to use larger model!
tokens = nlp("dog cat banana")

for token1 in tokens:
    for token2 in tokens:
        print(token1.text, token2.text, token1.similarity(token2))

In this case, the model’s predictions are pretty on point. A dog is very similar to a cat, whereas a banana is not very similar to either of them. Identical tokens are obviously 100% similar to each other (just not always exactly 1.0, because of vector math and floating point imprecisions).


When you call nlp on a text, spaCy first tokenizes the text to produce a Doc object. The Doc is then processed in several different steps – this is also referred to as the processing pipeline. The pipeline used by the default models consists of a tagger, a parser and an entity recognizer. Each pipeline component returns the processed Doc, which is then passed on to the next component.

The processing pipeline
tokenizerTokenizerDocSegment text into tokens.
taggerTaggerDoc[i].tagAssign part-of-speech tags.
parserDependencyParserDoc[i].head, Doc[i].dep, Doc.sents, Doc.noun_chunksAssign dependency labels.
nerEntityRecognizerDoc.ents, Doc[i].ent_iob, Doc[i].ent_typeDetect and label named entities.
textcatTextCategorizerDoc.catsAssign document labels.
custom,, Span._.xxxAssign custom attributes, methods or properties.

The processing pipeline always depends on the statistical model and its capabilities. For example, a pipeline can only include an entity recognizer component if the model includes data to make predictions of entity labels. This is why each model will specify the pipeline to use in its meta data, as a simple list containing the component names:

"pipeline": ["tagger", "parser", "ner"]

In spaCy v2.x, the statistical components like the tagger or parser are independent and don’t share any data between themselves. For example, the named entity recognizer doesn’t use any features set by the tagger and parser, and so on. This means that you can swap them, or remove single components from the pipeline without affecting the others.

However, custom components may depend on annotations set by other components. For example, a custom lemmatizer may need the part-of-speech tags assigned, so it’ll only work if it’s added after the tagger. The parser will respect pre-defined sentence boundaries, so if a previous component in the pipeline sets them, its dependency predictions may be different. Similarly, it matters if you add the EntityRuler before or after the statistical entity recognizer: if it’s added before, the entity recognizer will take the existing entities into account when making predictions. The EntityLinker, which resolves named entities to knowledge base IDs, should be preceded by a pipeline component that recognizes entities such as the EntityRecognizer.

The tokenizer is a “special” component and isn’t part of the regular pipeline. It also doesn’t show up in nlp.pipe_names. The reason is that there can only really be one tokenizer, and while all other pipeline components take a Doc and return it, the tokenizer takes a string of text and turns it into a Doc. You can still customize the tokenizer, though. nlp.tokenizer is writable, so you can either create your own Tokenizer class from scratch, or even replace it with an entirely custom function.

Vocab, hashes and lexemes

Whenever possible, spaCy tries to store data in a vocabulary, the Vocab, that will be shared by multiple documents. To save memory, spaCy also encodes all strings to hash values – in this case for example, “coffee” has the hash 3197928453018144401. Entity labels like “ORG” and part-of-speech tags like “VERB” are also encoded. Internally, spaCy only “speaks” in hash values.

Doc, Vocab, Lexeme and StringStore

If you process lots of documents containing the word “coffee” in all kinds of different contexts, storing the exact string “coffee” every time would take up way too much space. So instead, spaCy hashes the string and stores it in the StringStore. You can think of the StringStore as a lookup table that works in both directions – you can look up a string to get its hash, or a hash to get its string:

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("I love coffee")
print(doc.vocab.strings["coffee"])  # 3197928453018144401
print(doc.vocab.strings[3197928453018144401])  # 'coffee'

Now that all strings are encoded, the entries in the vocabulary don’t need to include the word text themselves. Instead, they can look it up in the StringStore via its hash value. Each entry in the vocabulary, also called Lexeme, contains the context-independent information about a word. For example, no matter if “love” is used as a verb or a noun in some context, its spelling and whether it consists of alphabetic characters won’t ever change. Its hash value will also always be the same.

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("I love coffee")
for word in doc:
    lexeme = doc.vocab[word.text]
    print(lexeme.text, lexeme.orth, lexeme.shape_, lexeme.prefix_, lexeme.suffix_,
            lexeme.is_alpha, lexeme.is_digit, lexeme.is_title, lexeme.lang_)

The mapping of words to hashes doesn’t depend on any state. To make sure each value is unique, spaCy uses a hash function to calculate the hash based on the word string. This also means that the hash for “coffee” will always be the same, no matter which model you’re using or how you’ve configured spaCy.

However, hashes cannot be reversed and there’s no way to resolve 3197928453018144401 back to “coffee”. All spaCy can do is look it up in the vocabulary. That’s why you always need to make sure all objects you create have access to the same vocabulary. If they don’t, spaCy might not be able to find the strings it needs.

import spacy
from spacy.tokens import Doc
from spacy.vocab import Vocab

nlp = spacy.load("en_core_web_sm")
doc = nlp("I love coffee")  # Original Doc
print(doc.vocab.strings["coffee"])  # 3197928453018144401
print(doc.vocab.strings[3197928453018144401])  # 'coffee' 👍

empty_doc = Doc(Vocab())  # New Doc with empty Vocab
# empty_doc.vocab.strings[3197928453018144401] will raise an error :(

empty_doc.vocab.strings.add("coffee")  # Add "coffee" and generate hash
print(empty_doc.vocab.strings[3197928453018144401])  # 'coffee' 👍

new_doc = Doc(doc.vocab)  # Create new doc with first doc's vocab
print(new_doc.vocab.strings[3197928453018144401])  # 'coffee' 👍

If the vocabulary doesn’t contain a string for 3197928453018144401, spaCy will raise an error. You can re-add “coffee” manually, but this only works if you actually know that the document contains that word. To prevent this problem, spaCy will also export the Vocab when you save a Doc or nlp object. This will give you the object and its encoded annotations, plus the “key” to decode it.

Knowledge Base

To support the entity linking task, spaCy stores external knowledge in a KnowledgeBase. The knowledge base (KB) uses the Vocab to store its data efficiently.

A knowledge base is created by first adding all entities to it. Next, for each potential mention or alias, a list of relevant KB IDs and their prior probabilities is added. The sum of these prior probabilities should never exceed 1 for any given alias.

import spacy
from spacy.kb import KnowledgeBase

# load the model and create an empty KB
nlp = spacy.load('en_core_web_sm')
kb = KnowledgeBase(vocab=nlp.vocab, entity_vector_length=3)

# adding entities
kb.add_entity(entity="Q1004791", freq=6, entity_vector=[0, 3, 5])
kb.add_entity(entity="Q42", freq=342, entity_vector=[1, 9, -3])
kb.add_entity(entity="Q5301561", freq=12, entity_vector=[-2, 4, 2])

# adding aliases
kb.add_alias(alias="Douglas", entities=["Q1004791", "Q42", "Q5301561"], probabilities=[0.6, 0.1, 0.2])
kb.add_alias(alias="Douglas Adams", entities=["Q42"], probabilities=[0.9])

print("Number of entities in KB:",kb.get_size_entities()) # 3
print("Number of aliases in KB:", kb.get_size_aliases()) # 2

Candidate generation

Given a textual entity, the Knowledge Base can provide a list of plausible candidates or entity identifiers. The EntityLinker will take this list of candidates as input, and disambiguate the mention to the most probable identifier, given the document context.

import spacy
from spacy.kb import KnowledgeBase

nlp = spacy.load('en_core_web_sm')
kb = KnowledgeBase(vocab=nlp.vocab, entity_vector_length=3)

# adding entities
kb.add_entity(entity="Q1004791", freq=6, entity_vector=[0, 3, 5])
kb.add_entity(entity="Q42", freq=342, entity_vector=[1, 9, -3])
kb.add_entity(entity="Q5301561", freq=12, entity_vector=[-2, 4, 2])

# adding aliases
kb.add_alias(alias="Douglas", entities=["Q1004791", "Q42", "Q5301561"], probabilities=[0.6, 0.1, 0.2])

candidates = kb.get_candidates("Douglas")
for c in candidates:
    print(" ", c.entity_, c.prior_prob, c.entity_vector)


If you’ve been modifying the pipeline, vocabulary, vectors and entities, or made updates to the model, you’ll eventually want to save your progress – for example, everything that’s in your nlp object. This means you’ll have to translate its contents and structure into a format that can be saved, like a file or a byte string. This process is called serialization. spaCy comes with built-in serialization methods and supports the Pickle protocol.

All container classes, i.e. Language (nlp), Doc, Vocab and StringStore have the following methods available:

to_bytesbytesdata = nlp.to_bytes()


spaCy’s models are statistical and every “decision” they make – for example, which part-of-speech tag to assign, or whether a word is a named entity – is a prediction. This prediction is based on the examples the model has seen during training. To train a model, you first need training data – examples of text, and the labels you want the model to predict. This could be a part-of-speech tag, a named entity or any other information.

The model is then shown the unlabelled text and will make a prediction. Because we know the correct answer, we can give the model feedback on its prediction in the form of an error gradient of the loss function that calculates the difference between the training example and the expected output. The greater the difference, the more significant the gradient and the updates to our model.

The training process

When training a model, we don’t just want it to memorize our examples – we want it to come up with a theory that can be generalized across other examples. After all, we don’t just want the model to learn that this one instance of “Amazon” right here is a company – we want it to learn that “Amazon”, in contexts like this, is most likely a company. That’s why the training data should always be representative of the data we want to process. A model trained on Wikipedia, where sentences in the first person are extremely rare, will likely perform badly on Twitter. Similarly, a model trained on romantic novels will likely perform badly on legal text.

This also means that in order to know how the model is performing, and whether it’s learning the right things, you don’t only need training data – you’ll also need evaluation data. If you only test the model with the data it was trained on, you’ll have no idea how well it’s generalizing. If you want to train a model from scratch, you usually need at least a few hundred examples for both training and evaluation. To update an existing model, you can already achieve decent results with very few examples – as long as they’re representative.

Language data

Every language is different – and usually full of exceptions and special cases, especially amongst the most common words. Some of these exceptions are shared across languages, while others are entirely specific – usually so specific that they need to be hard-coded. The lang module contains all language-specific data, organized in simple Python files. This makes the data easy to update and extend.

The shared language data in the directory root includes rules that can be generalized across languages – for example, rules for basic punctuation, emoji, emoticons, single-letter abbreviations and norms for equivalent tokens with different spellings, like " and . This helps the models make more accurate predictions. The individual language data in a submodule contains rules that are only relevant to a particular language. It also takes care of putting together all components and creating the Language subclass – for example, English or German.

Language data architecture
Stop words
List of most common words of a language that are often useful to filter out, for example “and” or “I”. Matching tokens will return True for is_stop.
Tokenizer exceptions
Special-case rules for the tokenizer, for example, contractions like “can’t” and abbreviations with punctuation, like “U.K.”.
Norm exceptions
Special-case rules for normalizing tokens to improve the model’s predictions, for example on American vs. British spelling.
Punctuation rules
Regular expressions for splitting tokens, e.g. on punctuation or special characters like emoji. Includes rules for prefixes, suffixes and infixes.
Character classes
Character classes to be used in regular expressions, for example, latin characters, quotes, hyphens or icons.
Lexical attributes
Custom functions for setting lexical attributes on tokens, e.g. like_num, which includes language-specific words like “ten” or “hundred”.
Syntax iterators
Functions that compute views of a Doc object based on its syntax. At the moment, only used for noun chunks.
Tag map
Dictionary mapping strings in your tag set to Universal Dependencies tags.
Morph rules
Exception rules for morphological analysis of irregular words like personal pronouns.
Lemmatization rules or a lookup-based lemmatization table to assign base forms, for example “be” for “was”.

Lightning tour

The following examples and code snippets give you an overview of spaCy’s functionality and its usage.

Install models and process text

python -m spacy download en_core_web_sm
python -m spacy download de_core_news_sm
import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("Hello, world. Here are two sentences.")
print([t.text for t in doc])

nlp_de = spacy.load("de_core_news_sm")
doc_de = nlp_de("Ich bin ein Berliner.")
print([t.text for t in doc_de])

Get tokens, noun chunks & sentences Needs model

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("Peach emoji is where it has always been. Peach is the superior "
          "emoji. It's outranking eggplant 🍑 ")
print(doc[0].text)          # 'Peach'
print(doc[1].text)          # 'emoji'
print(doc[-1].text)         # '🍑'
print(doc[17:19].text)      # 'outranking eggplant'

noun_chunks = list(doc.noun_chunks)
print(noun_chunks[0].text)  # 'Peach emoji'

sentences = list(doc.sents)
assert len(sentences) == 3
print(sentences[1].text)    # 'Peach is the superior emoji.'

Get part-of-speech tags and flags Needs model

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("Apple is looking at buying U.K. startup for $1 billion")
apple = doc[0]
print("Fine-grained POS tag", apple.pos_, apple.pos)
print("Coarse-grained POS tag", apple.tag_, apple.tag)
print("Word shape", apple.shape_, apple.shape)
print("Alphabetic characters?", apple.is_alpha)
print("Punctuation mark?", apple.is_punct)

billion = doc[10]
print("Digit?", billion.is_digit)
print("Like a number?", billion.like_num)
print("Like an email address?", billion.like_email)

Use hash values for any string

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("I love coffee")

coffee_hash = nlp.vocab.strings["coffee"]  # 3197928453018144401
coffee_text = nlp.vocab.strings[coffee_hash]  # 'coffee'
print(coffee_hash, coffee_text)
print(doc[2].orth, coffee_hash)  # 3197928453018144401
print(doc[2].text, coffee_text)  # 'coffee'

beer_hash = doc.vocab.strings.add("beer")  # 3073001599257881079
beer_text = doc.vocab.strings[beer_hash]  # 'beer'
print(beer_hash, beer_text)

unicorn_hash = doc.vocab.strings.add("🦄")  # 18234233413267120783
unicorn_text = doc.vocab.strings[unicorn_hash]  # '🦄'
print(unicorn_hash, unicorn_text)

Recognize and update named entities Needs model

import spacy
from spacy.tokens import Span

nlp = spacy.load("en_core_web_sm")
doc = nlp("San Francisco considers banning sidewalk delivery robots")
for ent in doc.ents:
    print(ent.text, ent.start_char, ent.end_char, ent.label_)

doc = nlp("FB is hiring a new VP of global policy")
doc.ents = [Span(doc, 0, 1, label="ORG")]
for ent in doc.ents:
    print(ent.text, ent.start_char, ent.end_char, ent.label_)

Train and update neural network models

import spacy
import random

nlp = spacy.load("en_core_web_sm")
train_data = [("Uber blew through $1 million", {"entities": [(0, 4, "ORG")]})]

other_pipes = [pipe for pipe in nlp.pipe_names if pipe != "ner"]
with nlp.disable_pipes(*other_pipes):
    optimizer = nlp.begin_training()
    for i in range(10):
        for text, annotations in train_data:
            nlp.update([text], [annotations], sgd=optimizer)

Visualize a dependency parse and named entities in your browser v2.0Needs model

from spacy import displacy

doc_dep = nlp("This is a sentence.")
displacy.serve(doc_dep, style="dep")

doc_ent = nlp("When Sebastian Thrun started working on self-driving cars at Google "
              "in 2007, few people outside of the company took him seriously.")
displacy.serve(doc_ent, style="ent")

Get word vectors and similarity Needs model

import spacy

nlp = spacy.load("en_core_web_md")
doc = nlp("Apple and banana are similar. Pasta and hippo aren't.")

apple = doc[0]
banana = doc[2]
pasta = doc[6]
hippo = doc[8]

print("apple <-> banana", apple.similarity(banana))
print("pasta <-> hippo", pasta.similarity(hippo))
print(apple.has_vector, banana.has_vector, pasta.has_vector, hippo.has_vector)

For the best results, you should run this example using the en_vectors_web_lg model (currently not available in the live demo).

Simple and efficient serialization

import spacy
from spacy.tokens import Doc
from spacy.vocab import Vocab

nlp = spacy.load("en_core_web_sm")
customer_feedback = open("customer_feedback_627.txt").read()
doc = nlp(customer_feedback)

new_doc = Doc(Vocab()).from_disk("/tmp/customer_feedback_627.bin")

Match text with token rules

import spacy
from spacy.matcher import Matcher

nlp = spacy.load("en_core_web_sm")
matcher = Matcher(nlp.vocab)

def set_sentiment(matcher, doc, i, matches):
    doc.sentiment += 0.1

pattern1 = [{"ORTH": "Google"}, {"ORTH": "I"}, {"ORTH": "/"}, {"ORTH": "O"}]
pattern2 = [[{"ORTH": emoji, "OP": "+"}] for emoji in ["😀", "😂", "🤣", "😍"]]
matcher.add("GoogleIO", None, pattern1)  # Match "Google I/O" or "Google i/o"
matcher.add("HAPPY", set_sentiment, *pattern2)  # Match one or more happy emoji

doc = nlp("A text about Google I/O 😀😀")
matches = matcher(doc)

for match_id, start, end in matches:
    string_id = nlp.vocab.strings[match_id]
    span = doc[start:end]
    print(string_id, span.text)
print("Sentiment", doc.sentiment)

Minibatched stream processing

texts = ["One document.", "...", "Lots of documents"]
# .pipe streams input, and produces streaming output
iter_texts = (texts[i % 3] for i in range(100000000))
for i, doc in enumerate(nlp.pipe(iter_texts, batch_size=50)):
    assert doc.is_parsed
    if i == 100:

Get syntactic dependencies Needs model

import spacy

nlp = spacy.load("en_core_web_sm")
doc = nlp("When Sebastian Thrun started working on self-driving cars at Google "
          "in 2007, few people outside of the company took him seriously.")

dep_labels = []
for token in doc:
    while token.head != token:
        token = token.head

Export to numpy arrays

import spacy
from spacy.attrs import ORTH, LIKE_URL

nlp = spacy.load("en_core_web_sm")
doc = nlp("Check out")
for token in doc:
    print(token.text, token.orth, token.like_url)

attr_ids = [ORTH, LIKE_URL]
doc_array = doc.to_array(attr_ids)
print(len(doc), len(attr_ids))

assert doc[0].orth == doc_array[0, 0]
assert doc[1].orth == doc_array[1, 0]
assert doc[0].like_url == doc_array[0, 1]

assert list(doc_array[:, 1]) == [t.like_url for t in doc]
print(list(doc_array[:, 1]))

Calculate inline markup on original string

import spacy

def put_spans_around_tokens(doc):
    """Here, we're building a custom "syntax highlighter" for
    part-of-speech tags and dependencies. We put each token in a
    span element, with the appropriate classes computed. All whitespace is
    preserved, outside of the spans. (Of course, HTML will only display
    multiple whitespace if enabled – but the point is, no information is lost
    and you can calculate what you need, e.g. <br />, <p> etc.)
    output = []
    html = '<span class="{classes}">{word}</span>{space}'
    for token in doc:
        if token.is_space:
            classes = "pos-{} dep-{}".format(token.pos_, token.dep_)
            output.append(html.format(classes=classes, word=token.text, space=token.whitespace_))
    string = "".join(output)
    string = string.replace("\n", "")
    string = string.replace("\t", "    ")
    return "<pre>{}</pre>".format(string)

nlp = spacy.load("en_core_web_sm")
doc = nlp("This is a test.\n\nHello   world.")
html = put_spans_around_tokens(doc)


The central data structures in spaCy are the Doc and the Vocab. The Doc object owns the sequence of tokens and all their annotations. The Vocab object owns a set of look-up tables that make common information available across documents. By centralizing strings, word vectors and lexical attributes, we avoid storing multiple copies of this data. This saves memory, and ensures there’s a single source of truth.

Text annotations are also designed to allow a single source of truth: the Doc object owns the data, and Span and Token are views that point into it. The Doc object is constructed by the Tokenizer, and then modified in place by the components of the pipeline. The Language object coordinates these components. It takes raw text and sends it through the pipeline, returning an annotated document. It also orchestrates training and serialization.

Library architecture

Container objects

DocA container for accessing linguistic annotations.
SpanA slice from a Doc object.
TokenAn individual token — i.e. a word, punctuation symbol, whitespace, etc.
LexemeAn entry in the vocabulary. It’s a word type with no context, as opposed to a word token. It therefore has no part-of-speech tag, dependency parse etc.

Processing pipeline

LanguageA text-processing pipeline. Usually you’ll load this once per process as nlp and pass the instance around your application.
TokenizerSegment text, and create Doc objects with the discovered segment boundaries.
LemmatizerDetermine the base forms of words.
MorphologyAssign linguistic features like lemmas, noun case, verb tense etc. based on the word and its part-of-speech tag.
TaggerAnnotate part-of-speech tags on Doc objects.
DependencyParserAnnotate syntactic dependencies on Doc objects.
EntityRecognizerAnnotate named entities, e.g. persons or products, on Doc objects.
TextCategorizerAssign categories or labels to Doc objects.
MatcherMatch sequences of tokens, based on pattern rules, similar to regular expressions.
PhraseMatcherMatch sequences of tokens based on phrases.
EntityRulerAdd entity spans to the Doc using token-based rules or exact phrase matches.
SentencizerImplement custom sentence boundary detection logic that doesn’t require the dependency parse.
Other functionsAutomatically apply something to the Doc, e.g. to merge spans of tokens.

Other classes

VocabA lookup table for the vocabulary that allows you to access Lexeme objects.
StringStoreMap strings to and from hash values.
VectorsContainer class for vector data keyed by string.
GoldParseCollection for training annotations.
GoldCorpusAn annotated corpus, using the JSON file format. Manages annotations for tagging, dependency parsing and NER.

Community & FAQ

We’re very happy to see the spaCy community grow and include a mix of people from all kinds of different backgrounds – computational linguistics, data science, deep learning, research and more. If you’d like to get involved, below are some answers to the most important questions and resources for further reading.

Help, my code isn’t working!

Bugs suck, and we’re doing our best to continuously improve the tests and fix bugs as soon as possible. Before you submit an issue, do a quick search and check if the problem has already been reported. If you’re having installation or loading problems, make sure to also check out the troubleshooting guide. Help with spaCy is available via the following platforms:

  • Stack Overflow: Usage questions and everything related to problems with your specific code. The Stack Overflow community is much larger than ours, so if your problem can be solved by others, you’ll receive help much quicker.
  • GitHub discussions: General discussion, project ideas and usage questions. Meet other community members to get help with a specific code implementation, discuss ideas for new projects/plugins, support more languages, and share best practices.
  • GitHub issue tracker: Bug reports and improvement suggestions, i.e. everything that’s likely spaCy’s fault. This also includes problems with the models beyond statistical imprecisions, like patterns that point to a bug.

How can I contribute to spaCy?

You don’t have to be an NLP expert or Python pro to contribute, and we’re happy to help you get started. If you’re new to spaCy, a good place to start is the help wanted (easy) label on GitHub, which we use to tag bugs and feature requests that are easy and self-contained. We also appreciate contributions to the docs – whether it’s fixing a typo, improving an example or adding additional explanations. You’ll find a “Suggest edits” link at the bottom of each page that points you to the source.

Another way of getting involved is to help us improve the language data – especially if you happen to speak one of the languages currently in alpha support. Even adding simple tokenizer exceptions, stop words or lemmatizer data can make a big difference. It will also make it easier for us to provide a statistical model for the language in the future. Submitting a test that documents a bug or performance issue, or covers functionality that’s especially important for your application is also very helpful. This way, you’ll also make sure we never accidentally introduce regressions to the parts of the library that you care about the most.

For more details on the types of contributions we’re looking for, the code conventions and other useful tips, make sure to check out the contributing guidelines.

I’ve built something cool with spaCy – how can I get the word out?

First, congrats – we’d love to check it out! When you share your project on Twitter, don’t forget to tag @spacy_io so we don’t miss it. If you think your project would be a good fit for the spaCy Universe, feel free to submit it! Tutorials are also incredibly valuable to other users and a great way to get exposure. So we strongly encourage writing up your experiences, or sharing your code and some tips and tricks on your blog. Since our website is open-source, you can add your project or tutorial by making a pull request on GitHub.

If you would like to use the spaCy logo on your site, please get in touch and ask us first. However, if you want to show support and tell others that your project is using spaCy, you can grab one of our spaCy badges here:

[![Built with spaCy](](
[![Built with spaCy](❤%20and-spaCy-09a3d5.svg)](