tests/testthat/test-transformers.R

In farach/huggingfaceR: Hugging Face State-of-the-Art Models in R

test_that("load model fails if not given correct model nae", {
  expect_error(hf_load_pipeline(model_id = "x not a model x"))
})

#Tokenizer tests
tokenizer <- hf_load_tokenizer("distilbert-base-uncased")
slow_tokenizer <- hf_load_tokenizer('distilbert-base-uncased', use_fast = FALSE)

test_that("incorrect model not loaded", {
  expect_false(stringr::str_detect(as.character(tokenizer), "GPT|gpt"))
})

test_that("tokenizer loads correct model", {
  expect_true(stringr::str_detect(as.character(tokenizer),"distilbert"))
})

test_that("load tokenizer function loads correctly", {
  expect_true("vocab" %in% names(tokenizer))
})

test_that("tokenizer function properly passess ...",{
  expect_false(stringr::str_detect(as.character(slow_tokenizer), "is_fast=True"))
})

test_that("tokenizer loads fast if not told otherwise by ...",{
  expect_true(stringr::str_detect(as.character(tokenizer), "is_fast=True"))
})


#placeholder text for AutoModelForX tests
qa_model <- hf_load_AutoModel_for_task(model_type = "AutoModelForQuestionAnswering",
                              model_id = "deepset/roberta-base-squad2")

test_that("Question Answering Model is of correct type", {
  expect_true(stringr::str_detect(as.character(qa_model$config), "QuestionAnswering"))
})

sent_model <- hf_load_AutoModel_for_task(model_id = "cardiffnlp/twitter-roberta-base-sentiment-latest",
                  model_type = "AutoModelForSequenceClassification")

test_that("Sentiment model is of Sequence Classification type", {
  expect_true(stringr::str_detect(as.character(sent_model$config), "ForSequenceClass"))
})


test_that("Sentiment model outputs 3 classes", {
  expect_equal(sent_model$config$num_labels, 3)
})

test_that("Sentiment model has label2id and id2label", {
  expect_true("label2id" %in% names(sent_model$config))
  expect_true("id2label" %in% names(sent_model$config))
})

#pipeline tests
pipe <- hf_load_pipeline(model_id = "distilbert-base-uncased")
pipe_qa <- hf_load_pipeline(model_id = "distilbert-base-uncased", task = "question-answering")

test_that("hf_load_pipeline loads tokenizer", {
  expect_true(stringr::str_detect(as.character(pipe$tokenizer), "distilb"))
})

test_that("hf_load_pipeline does not load incorrect model", {
  expect_false(stringr::str_detect(as.character(pipe$model), "GPT|gpt"))
})

test_that("hf_load_pipeline loads correct model", {
  expect_true(stringr::str_detect(as.character(pipe$model), "distilb"))
})

test_that("hf_load_pipeline task argument is functioning", {

  expect_equal("question-answering", pipe_qa$task)
})