Train static word embeddings using the Word2Vec, GloVe, or FastText algorithm.

Train static word embeddings using the Word2Vec, GloVe, or FastText algorithm with multi-threading.

Usage

train_wordvec(
  text,
  method = c("word2vec", "glove", "fasttext"),
  dims = 300,
  window = 5,
  min.freq = 5,
  threads = 8,
  model = c("skip-gram", "cbow"),
  loss = c("ns", "hs"),
  negative = 5,
  subsample = 1e-04,
  learning = 0.05,
  ngrams = c(3, 6),
  x.max = 10,
  convergence = -1,
  stopwords = character(0),
  encoding = "UTF-8",
  tolower = FALSE,
  normalize = FALSE,
  iteration,
  tokenizer,
  remove,
  file.save,
  compress = "bzip2",
  verbose = TRUE
)

Arguments

text

A character vector of text, or a file path on disk containing text.

method

Training algorithm:

• "word2vec" (default): using word2vec::word2vec()

• "glove": using rsparse::GloVe() and text2vec::text2vec()

• "fasttext": using fastTextR::ft_train()

dims

Number of dimensions of word vectors to be trained. Common choices include 50, 100, 200, 300, and 500. Defaults to 300.

window

Window size (number of nearby words behind/ahead the current word). It defines how many surrounding words to be included in training: [window] words behind and [window] words ahead ([window]*2 in total). Defaults to 5.

min.freq

Minimum frequency of words to be included in training. Words that appear less than this value of times will be excluded from vocabulary. Defaults to 5 (take words that appear at least five times).

threads

Number of CPU threads used for training. A modest value produces the fastest training. Too many threads are not always helpful. Defaults to 8.

model

[Only for Word2Vec / FastText] Learning model architecture:

• "skip-gram" (default): Skip-Gram, which predicts surrounding words given the current word

• "cbow": Continuous Bag-of-Words, which predicts the current word based on the context

loss

[Only for Word2Vec / FastText] Loss function (computationally efficient approximation):

• "ns" (default): Negative Sampling

• "hs": Hierarchical Softmax

negative

[Only for Negative Sampling in Word2Vec / FastText] Number of negative examples. Values in the range 5~20 are useful for small training datasets, while for large datasets the value can be as small as 2~5. Defaults to 5.

subsample

[Only for Word2Vec / FastText] Subsampling of frequent words (threshold for occurrence of words). Those that appear with higher frequency in the training data will be randomly down-sampled. Defaults to 0.0001 (1e-04).

learning

[Only for Word2Vec / FastText] Initial (starting) learning rate, also known as alpha. Defaults to 0.05.

ngrams

[Only for FastText] Minimal and maximal ngram length. Defaults to c(3, 6).

x.max

[Only for GloVe] Maximum number of co-occurrences to use in the weighting function. Defaults to 10.

convergence

[Only for GloVe] Convergence tolerance for SGD iterations. Defaults to -1.

stopwords

[Only for Word2Vec / GloVe] A character vector of stopwords to be excluded from training.

encoding

Text encoding. Defaults to "UTF-8".

tolower

Convert all upper-case characters to lower-case? Defaults to FALSE.

normalize

Normalize all word vectors to unit length? Defaults to FALSE. See normalize().

iteration

Number of training iterations. More iterations makes a more precise model, but computational cost is linearly proportional to iterations. Defaults to 5 for Word2Vec and FastText while 10 for GloVe.

tokenizer

Function used to tokenize the text. Defaults to text2vec::word_tokenizer().

remove

Strings (in regular expression) to be removed from the text. Defaults to "_|'|<br/>|<br />|e\\\\.g\\\\.|i\\\\.e\\\\.". You may turn off this by specifying remove=NULL.

file.save

File name of to-be-saved R data (must be .RData).

compress

Compression method for the saved file. Defaults to "bzip2".

• 1 or "gzip": modest file size (fastest)

• 2 or "bzip2": small file size (fast)

• 3 or "xz": minimized file size (slow)

verbose

Print information to the console? Defaults to TRUE.

Value

A wordvec (data.table) with three variables: word, vec, freq.

Download

Download pre-trained word vectors data (.RData): https://psychbruce.github.io/WordVector_RData.pdf

References

All-in-one package:

• https://CRAN.R-project.org/package=wordsalad

Word2Vec:

• https://code.google.com/archive/p/word2vec/

• https://CRAN.R-project.org/package=word2vec

• https://github.com/maxoodf/word2vec

GloVe:

• https://nlp.stanford.edu/projects/glove/

• https://text2vec.org/glove.html

• https://CRAN.R-project.org/package=text2vec

• https://CRAN.R-project.org/package=rsparse

FastText:

• https://fasttext.cc/

• https://CRAN.R-project.org/package=fastTextR

See also

tokenize()

Examples

review = text2vec::movie_review  # a data.frame'
text = review$review

## Note: All the examples train 50 dims for faster code check.

## Word2Vec (SGNS)
dt1 = train_wordvec(
  text,
  method="word2vec",
  model="skip-gram",
  dims=50, window=5,
  normalize=TRUE)
#> ✔ Tokenized: 70105 sentences (time cost = 2 secs)
#> ✔ Text corpus: 5242249 characters, 1185427 tokens (roughly words)
#> 
#> ── Training model information ──────────────────────────────────────────────────
#> - Method:      Word2Vec (Skip-Gram with Negative Sampling)
#> - Dimensions:  50
#> - Window size: 5 (5 words behind and 5 words ahead the current word)
#> - Subsampling: 1e-04
#> - Min. freq.:  5 occurrences in text
#> - Iterations:  5 training iterations
#> - CPU threads: 8
#> 
#> ── Training... 
#> ✔ Word vectors trained: 14205 unique tokens (time cost = 5 secs)

dt1
#> # wordvec (data.table): [14205 × 3] (normalized)
#>          word                     vec  freq
#>     1:    the [ 0.1920, ...<50 dims>] 58797
#>     2:    and [ 0.0153, ...<50 dims>] 32193
#>     3:      a [ 0.0948, ...<50 dims>] 31783
#>     4:     of [-0.0234, ...<50 dims>] 29142
#>     5:     to [ 0.1207, ...<50 dims>] 27218
#> ------                                     
#> 14201: drunks [ 0.1232, ...<50 dims>]     5
#> 14202:   flea [ 0.1208, ...<50 dims>]     5
#> 14203: liquid [ 0.1434, ...<50 dims>]     5
#> 14204:   LOTR [ 0.1306, ...<50 dims>]     5
#> 14205: morose [ 0.1506, ...<50 dims>]     5
most_similar(dt1, "Ive")  # evaluate performance
#> [Word Vector] =~ Ive
#> (normalized to unit length)
#>           word   cos_sim row_id
#>         <char>     <num>  <int>
#>  1:        ive 0.8877912    110
#>  2:       seen 0.8031776    766
#>  3:     havent 0.7913808    943
#>  4: criticized 0.7844135   1131
#>  5:  Bollywood 0.7789753   3127
#>  6:        dub 0.7767296   3487
#>  7:      youve 0.7697009   3586
#>  8:      WORST 0.7686428   4821
#>  9:      quote 0.7680053   6904
#> 10:     recent 0.7652648   8453
most_similar(dt1, ~ man - he + she, topn=5)  # evaluate performance
#> [Word Vector] =~ man - he + she
#> (normalized to unit length)
#>       word   cos_sim row_id
#>     <char>     <num>  <int>
#> 1:   woman 0.8964092    260
#> 2:    girl 0.8416096    299
#> 3:  lonely 0.7838781    478
#> 4:     boy 0.7815670   1213
#> 5: married 0.7659517   2338
most_similar(dt1, ~ boy - he + she, topn=5)  # evaluate performance
#> [Word Vector] =~ boy - he + she
#> (normalized to unit length)
#>      word   cos_sim row_id
#>    <char>     <num>  <int>
#> 1:   girl 0.8561397    197
#> 2:  woman 0.7813664    260
#> 3:   aged 0.7309787    299
#> 4:  young 0.7034726   1140
#> 5:   lady 0.6838961   1836

## GloVe
dt2 = train_wordvec(
  text,
  method="glove",
  dims=50, window=5,
  normalize=TRUE)
#> ✔ Tokenized: 70105 sentences (time cost = 2 secs)
#> ✔ Text corpus: 5242249 characters, 1185427 tokens (roughly words)
#> 
#> ── Training model information ──────────────────────────────────────────────────
#> - Method:      GloVe
#> - Dimensions:  50
#> - Window size: 5 (5 words behind and 5 words ahead the current word)
#> - Subsampling: N/A
#> - Min. freq.:  5 occurrences in text
#> - Iterations:  10 training iterations
#> - CPU threads: 8
#> 
#> ── Training... 
#> ✔ Word vectors trained: 14207 unique tokens (time cost = 10 secs)

dt2
#> # wordvec (data.table): [14207 × 3] (normalized)
#>            word                     vec  freq
#>     1:      the [ 0.1325, ...<50 dims>] 58797
#>     2:      and [ 0.0807, ...<50 dims>] 32193
#>     3:        a [ 0.1503, ...<50 dims>] 31783
#>     4:       of [-0.1085, ...<50 dims>] 29142
#>     5:       to [ 0.1437, ...<50 dims>] 27218
#> ------                                       
#> 14203:      yea [ 0.1552, ...<50 dims>]     5
#> 14204:   yearly [ 0.0437, ...<50 dims>]     5
#> 14205: yearning [-0.0816, ...<50 dims>]     5
#> 14206:   yelled [-0.1324, ...<50 dims>]     5
#> 14207:      yer [-0.0856, ...<50 dims>]     5
most_similar(dt2, "Ive")  # evaluate performance
#> [Word Vector] =~ Ive
#> (normalized to unit length)
#>        word   cos_sim row_id
#>      <char>     <num>  <int>
#>  1:    seen 0.9229199     24
#>  2:    ever 0.8886434     91
#>  3:   worst 0.7852577    110
#>  4:   heard 0.7593296    124
#>  5:   since 0.7340906    261
#>  6: watched 0.7040197    262
#>  7:    read 0.6937020    305
#>  8:  movies 0.6933035    330
#>  9: already 0.6910354    468
#> 10:    have 0.6736166    515
most_similar(dt2, ~ man - he + she, topn=5)  # evaluate performance
#> [Word Vector] =~ man - he + she
#> (normalized to unit length)
#>      word   cos_sim row_id
#>    <char>     <num>  <int>
#> 1:  woman 0.8647834     34
#> 2:  young 0.7469375    198
#> 3:  child 0.7445291    260
#> 4:   girl 0.7283451    299
#> 5:    who 0.7177403    523
most_similar(dt2, ~ boy - he + she, topn=5)  # evaluate performance
#> [Word Vector] =~ boy - he + she
#> (normalized to unit length)
#>      word   cos_sim row_id
#>    <char>     <num>  <int>
#> 1:   girl 0.8038970    150
#> 2:  woman 0.7168044    198
#> 3:  young 0.7102703    260
#> 4:  named 0.7032806    299
#> 5:    old 0.6642969    867

## FastText
dt3 = train_wordvec(
  text,
  method="fasttext",
  model="skip-gram",
  dims=50, window=5,
  normalize=TRUE)
#> ✔ Tokenized: 70105 sentences (time cost = 2 secs)
#> ✔ Text corpus: 5242249 characters, 1185427 tokens (roughly words)
#> 
#> ── Training model information ──────────────────────────────────────────────────
#> - Method:      FastText (Skip-Gram with Negative Sampling)
#> - Dimensions:  50
#> - Window size: 5 (5 words behind and 5 words ahead the current word)
#> - Subsampling: 1e-04
#> - Min. freq.:  5 occurrences in text
#> - Iterations:  5 training iterations
#> - CPU threads: 8
#> 
#> ── Training... 
#> ✔ Word vectors trained: 14207 unique tokens (time cost = 12 secs)

dt3
#> # wordvec (data.table): [14207 × 3] (normalized)
#>                word                     vec  freq
#>     1:          the [ 0.0145, ...<50 dims>] 58797
#>     2:          and [-0.0386, ...<50 dims>] 32193
#>     3:            a [ 0.0273, ...<50 dims>] 31783
#>     4:           of [ 0.0613, ...<50 dims>] 29142
#>     5:           to [-0.0866, ...<50 dims>] 27218
#> ------                                           
#> 14203:        spray [ 0.0716, ...<50 dims>]     5
#> 14204: disabilities [ 0.1356, ...<50 dims>]     5
#> 14205:        crook [ 0.0305, ...<50 dims>]     5
#> 14206:     Syndrome [-0.0106, ...<50 dims>]     5
#> 14207:      snipers [ 0.0300, ...<50 dims>]     5
most_similar(dt3, "Ive")  # evaluate performance
#> [Word Vector] =~ Ive
#> (normalized to unit length)
#>        word   cos_sim row_id
#>      <char>     <num>  <int>
#>  1:   Youve 0.8561718    110
#>  2:    seen 0.8367139    945
#>  3:    Weve 0.8362379   2612
#>  4:   youve 0.8028984   3105
#>  5:   WORST 0.7753204   3494
#>  6:      ve 0.7745071   5898
#>  7:     ive 0.7565490   6913
#>  8: funnier 0.7524717   7108
#>  9:    weve 0.7469759   8617
#> 10: Daisies 0.7461700  12894
most_similar(dt3, ~ man - he + she, topn=5)  # evaluate performance
#> [Word Vector] =~ man - he + she
#> (normalized to unit length)
#>        word   cos_sim row_id
#>      <char>     <num>  <int>
#> 1:    woman 0.8977993    261
#> 2:     girl 0.7848848    299
#> 3: salesman 0.7776333   5594
#> 4:   madman 0.7481051   6553
#> 5: henchman 0.7462895  12263
most_similar(dt3, ~ boy - he + she, topn=5)  # evaluate performance
#> [Word Vector] =~ boy - he + she
#> (normalized to unit length)
#>      word   cos_sim row_id
#>    <char>     <num>  <int>
#> 1:   girl 0.7819794    261
#> 2:   boys 0.7061669    299
#> 3:  woman 0.6909817    523
#> 4:    kid 0.6832291    676
#> 5:  child 0.6823652   1045