State-of-the-art NLP fashions from R


The Transformers repository from “Hugging Face” incorporates plenty of prepared to make use of, state-of-the-art fashions, that are simple to obtain and fine-tune with Tensorflow & Keras.

For this objective the customers normally must get:

  • The mannequin itself (e.g. Bert, Albert, RoBerta, GPT-2 and and so forth.)
  • The tokenizer object
  • The weights of the mannequin

On this submit, we are going to work on a basic binary classification process and practice our dataset on 3 fashions:

Nevertheless, readers ought to know that one can work with transformers on a wide range of down-stream duties, akin to:

  1. characteristic extraction
  2. sentiment evaluation
  3. text classification
  4. question answering
  5. summarization
  6. translation and many more.


Our first job is to put in the transformers bundle through reticulate.

reticulate::py_install('transformers', pip = TRUE)

Then, as normal, load normal ‘Keras’, ‘TensorFlow’ >= 2.0 and a few basic libraries from R.

Be aware that if working TensorFlow on GPU one may specify the next parameters in an effort to keep away from reminiscence points.

physical_devices = tf$config$list_physical_devices('GPU')



We already talked about that to coach an information on the particular mannequin, customers ought to obtain the mannequin, its tokenizer object and weights. For instance, to get a RoBERTa mannequin one has to do the next:

# get Tokenizer
transformer$RobertaTokenizer$from_pretrained('roberta-base', do_lower_case=TRUE)

# get Mannequin with weights

Information preparation

A dataset for binary classification is offered in text2vec bundle. Let’s load the dataset and take a pattern for quick mannequin coaching.

Cut up our information into 2 elements:

idx_train =*0.8)

practice = df[idx_train,]
take a look at = df[!idx_train,]

Information enter for Keras

Till now, we’ve simply coated information import and train-test break up. To feed enter to the community we have now to show our uncooked textual content into indices through the imported tokenizer. After which adapt the mannequin to do binary classification by including a dense layer with a single unit on the finish.

Nevertheless, we wish to practice our information for 3 fashions GPT-2, RoBERTa, and Electra. We have to write a loop for that.

Be aware: one mannequin on the whole requires 500-700 MB

# listing of three fashions
ai_m = list(
  c('TFGPT2Model',       'GPT2Tokenizer',       'gpt2'),
   c('TFRobertaModel',    'RobertaTokenizer',    'roberta-base'),
   c('TFElectraModel',    'ElectraTokenizer',    'google/electra-small-generator')

# parameters
max_len = 50L
epochs = 2
batch_size = 10

# create an inventory for mannequin outcomes
gather_history = list()

for (i in 1:length(ai_m)) {
  # tokenizer
  tokenizer = glue::glue("transformer${ai_m[[i]][2]}$from_pretrained('{ai_m[[i]][3]}',
                         do_lower_case=TRUE)") %>% 
    rlang::parse_expr() %>% eval()
  # mannequin
  model_ = glue::glue("transformer${ai_m[[i]][1]}$from_pretrained('{ai_m[[i]][3]}')") %>% 
    rlang::parse_expr() %>% eval()
  # inputs
  textual content = list()
  # outputs
  label = list()
  data_prep = operate(information) {
    for (i in 1:nrow(information)) {
      txt = tokenizer$encode(information[['comment_text']][i],max_length = max_len, 
                             truncation=T) %>% 
        t() %>% 
        as.matrix() %>% list()
      lbl = information[['target']][i] %>% t()
      textual content = textual content %>% append(txt)
      label = label %>% append(lbl)
    list(,textual content),,label))
  train_ = data_prep(practice)
  test_ = data_prep(take a look at)
  # slice dataset
  tf_train = tensor_slices_dataset(list(train_[[1]],train_[[2]])) %>% 
    dataset_batch(batch_size = batch_size, drop_remainder = TRUE) %>% 
    dataset_shuffle(128) %>% dataset_repeat(epochs) %>% 
  tf_test = tensor_slices_dataset(list(test_[[1]],test_[[2]])) %>% 
    dataset_batch(batch_size = batch_size)
  # create an enter layer
  enter = layer_input(form=c(max_len), dtype='int32')
  hidden_mean = tf$reduce_mean(model_(enter)[[1]], axis=1L) %>% 
    layer_dense(64,activation = 'relu')
  # create an output layer for binary classification
  output = hidden_mean %>% layer_dense(models=1, activation='sigmoid')
  mannequin = keras_model(inputs=enter, outputs = output)
  # compile with AUC rating
  mannequin %>% compile(optimizer= tf$keras$optimizers$Adam(learning_rate=3e-5, epsilon=1e-08, clipnorm=1.0),
                    loss = tf$losses$BinaryCrossentropy(from_logits=F),
                    metrics = tf$metrics$AUC())
  # practice the mannequin
  historical past = mannequin %>% keras::match(tf_train, epochs=epochs, #steps_per_epoch=len/batch_size,
  gather_history[[i]]<- historical past
  names(gather_history)[i] = ai_m[[i]][1]

Reproduce in a           Notebook

Extract outcomes to see the benchmarks:

Each the RoBERTa and Electra fashions present some further enhancements after 2 epochs of coaching, which can’t be mentioned of GPT-2. On this case, it’s clear that it may be sufficient to coach a state-of-the-art mannequin even for a single epoch.


On this submit, we confirmed the best way to use state-of-the-art NLP fashions from R.
To know the best way to apply them to extra complicated duties, it’s extremely really useful to assessment the transformers tutorial.

We encourage readers to check out these fashions and share their outcomes beneath within the feedback part!


For those who see errors or wish to counsel modifications, please create an issue on the supply repository.


Textual content and figures are licensed below Artistic Commons Attribution CC BY 4.0. Supply code is accessible at, until in any other case famous. The figures which have been reused from different sources do not fall below this license and will be acknowledged by a word of their caption: “Determine from …”.


For attribution, please cite this work as

Abdullayev (2020, July 30). Posit AI Weblog: State-of-the-art NLP fashions from R. Retrieved from

BibTeX quotation

  creator = {Abdullayev, Turgut},
  title = {Posit AI Weblog: State-of-the-art NLP fashions from R},
  url = {},
  yr = {2020}

FNN-VAE for noisy time sequence forecasting

Parallelized sampling utilizing exponential variates