Edit on GitHub
  • John Snow Labs
  • Nov 03, 2025

CODER: Knowledge-Infused Biomedical Embedding Model (ONNX)

  • embeddings
  • en
  • biomedical
  • open_source
  • onnx

Description

This model is the English version of CODER, a knowledge-infused biomedical embedding model designed for medical concept normalization and cross-lingual representation learning.

Download Copy S3 URI

How to use

from sparknlp.base import *
from sparknlp.annotator import *
from pyspark.ml import Pipeline

documentAssembler = DocumentAssembler() \
    .setInputCol("text") \
    .setOutputCol("document")

tokenizer = Tokenizer() \
    .setInputCols("document") \
    .setOutputCol("token")

embeddings = BertEmbeddings.pretrained("umlsbert_eng_onnx", "en") \
    .setInputCols(["document", "token"]) \
    .setOutputCol("embeddings") \

pipeline = Pipeline(stages=[
    documentAssembler,
    tokenizer,
    embeddings
])

data = spark.createDataFrame([
    ["Artificial intelligence is transforming the world."],
    ["Machine learning enables powerful data-driven systems."]
]).toDF("text")

result = pipeline.fit(data).transform(data)
result.select("embeddings.embeddings").show()


import com.johnsnowlabs.nlp.base._
import com.johnsnowlabs.nlp.annotator._
import org.apache.spark.ml.Pipeline
import spark.implicits._

val documentAssembler = new DocumentAssembler()
  .setInputCol("text")
  .setOutputCol("document")

val tokenizer = new Tokenizer()
  .setInputCols("document")
  .setOutputCol("token")

val embeddings = BertEmbeddings
  .pretrained("umlsbert_eng_onnx", "en")
  .setInputCols(Array("document", "token"))
  .setOutputCol("embeddings")

val pipeline = new Pipeline().setStages(Array(
  documentAssembler,
  tokenizer,
  embeddings
))

val data = Seq(
  "Artificial intelligence is transforming the world.",
  "Machine learning enables powerful data-driven systems."
).toDF("text")

val result = pipeline.fit(data).transform(data)

result.select("embeddings.embeddings").show(truncate = false)

Results


+--------------------+
|          embeddings|
+--------------------+
|[[-0.6771237, 0.5...|
|[[-1.016453, 0.21...|
+--------------------+

Model Information

Model Name: umlsbert_eng_onnx
Compatibility: Healthcare NLP 6.1.0+
License: Open Source
Edition: Official
Input Labels: [document, token]
Output Labels: [embeddings]
Language: en
Size: 408.0 MB
Case sensitive: false
Max sentence length: 512