tests/testthat/testthat-chunk_entity_resolver.R

In r-spark/sparknlp: R Interface to John Snow Labs Spark NLP

setup({
  sc <- testthat_spark_connection()
  text_tbl <- testthat_tbl("test_text")

  train_data_file <- here::here("tests", "testthat", "data", "AskAPatient.fold-0.test.txt")
  train_data <- sparklyr::spark_read_csv(sc, "train", train_data_file, delimiter = "\t", 
                                         columns = c("conceptId", "_term", "term"))
  
  # These lines should set a pipeline that will ultimately create the columns needed for testing the annotator
  assembler <- nlp_document_assembler(sc, input_col = "term", output_col = "document")
  chunk <- nlp_doc2chunk(sc, input_cols = c("document"), output_col = "chunk")
  tokenizer <- nlp_tokenizer(sc, input_cols = c("document"), output_col = "token")
  embeddings <- nlp_word_embeddings_pretrained(sc, name = "embeddings_clinical",
                                               remote_loc = "clinical/models", 
                                               input_cols = c("document", "token"), output_col = "embeddings")
  chunk_emb <- nlp_chunk_embeddings(sc, input_cols = c("chunk", "embeddings"), output_col = "chunk_embeddings")

  pipeline <- ml_pipeline(assembler, chunk, tokenizer, embeddings, chunk_emb)
  test_data <- ml_fit_and_transform(pipeline, train_data)

  assign("sc", sc, envir = parent.frame())
  assign("pipeline", pipeline, envir = parent.frame())
  assign("test_data", test_data, envir = parent.frame())
})

teardown({
  rm(sc, envir = .GlobalEnv)
  rm(pipeline, envir = .GlobalEnv)
  rm(test_data, envir = .GlobalEnv)
})

test_that("chunk_entity_resolver param setting", {
  test_args <- list(
    input_cols = c("string1", "string2"),
    output_col = "string1",
    all_distances_metadata = TRUE,
    alternatives = 10,
    case_sensitive = FALSE,
    confidence_function = "string1",
    distance_function = "string1",
    distance_weights = c(0.6, 0.8),
    enable_jaccard = FALSE,
    enable_jaro_winkler = TRUE,
    enable_levenshtein = TRUE,
    enable_sorensen_dice = FALSE,
    enable_tfidf = FALSE,
    enable_wmd = TRUE,
    extra_mass_penalty = 0.3,
    label_column = NULL, # no getter or setter
    miss_as_empty = TRUE,
    neighbors = 5,
    normalized_col = "string1",
    pooling_strategy = "string1",
    threshold = 0.7
  )

  test_param_setting(sc, nlp_chunk_entity_resolver, test_args)
})

test_that("nlp_chunk_entity_resolver spark_connection", {
  test_annotator <- nlp_chunk_entity_resolver(sc, input_cols = c("token", "chunk_embeddings"),
                                              output_col = "recognized", label_column = "conceptId",
                                              normalized_col = "_term")
  fit_model <- ml_fit(test_annotator, test_data)
  transformed_data <- ml_transform(fit_model, test_data)
  expect_true("recognized" %in% colnames(transformed_data))

  expect_true(inherits(test_annotator, "nlp_chunk_entity_resolver"))
  expect_true(inherits(fit_model, "nlp_chunk_entity_resolver_model"))
})

test_that("nlp_chunk_entity_resolver ml_pipeline", {
  test_annotator <- nlp_chunk_entity_resolver(sc, input_cols = c("token", "chunk_embeddings"),
                                              output_col = "recognized", label_column = "conceptId",
                                              normalized_col = "_term")
  transformed_data <- ml_fit_and_transform(test_annotator, test_data)
  expect_true("recognized" %in% colnames(transformed_data))
})

test_that("nlp_chunk_entity_resolver tbl_spark", {
  transformed_data <- nlp_chunk_entity_resolver(test_data, input_cols = c("token", "chunk_embeddings"),
                                              output_col = "recognized", label_column = "conceptId",
                                              normalized_col = "_term")

  expect_true("recognized" %in% colnames(transformed_data))
})

test_that("nlp_chunk_entity_resolver pretrained", {
  model <- nlp_chunk_entity_resolver_pretrained(sc, input_cols = c("token", "chunk_embeddings"), 
                                                output_col = "recognized",
                                       name = "chunkresolve_snomed_findings_clinical", lang = "en", remote_loc = "clinical/models")
  transformed_data <- ml_transform(model, test_data)
  expect_true("recognized" %in% colnames(transformed_data))
  
  expect_true(inherits(model, "nlp_chunk_entity_resolver_model"))
})