Description
StarCoder2-3B model is a 3B parameter model trained on 17 programming languages from The Stack v2, with opt-out requests excluded. The model uses Grouped Query Attention, a context window of 16,384 tokens with a sliding window attention of 4,096 tokens, and was trained using the Fill-in-the-Middle objective on 3+ trillion tokens.
Predicted Entities
How to use
data = spark.createDataFrame([
[1, "def add(a, b):"]]).toDF("id", "text")
document_assembler = DocumentAssembler() \
.setInputCol("text") \
.setOutputCol("documents")
starcoder_loaded = StarCoderTransformer \
.pretrained("starcoder2_3b_int4","en") \
.setMaxOutputLength(50) \
.setDoSample(False) \
.setInputCols(["documents"]) \
.setOutputCol("generation")
pipeline = Pipeline().setStages([document_assembler, starcoder_loaded])
results = pipeline.fit(data).transform(data)
results.select("generation.result").show(truncate=False)
val documentAssembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val seq2seq = StarCoderTransformer.pretrained("starcoder2_3b_int4","en")
.setInputCols(Array("document"))
.setOutputCol("generation")
val pipeline = new Pipeline().setStages(Array(documentAssembler, seq2seq))
val data = Seq(""def add(a, b):").toDF("text")
val pipelineModel = pipeline.fit(data)
val pipelineDF = pipelineModel.transform(data)
Model Information
| Model Name: | starcoder2_3b_int4 |
| Compatibility: | Spark NLP 5.5.0+ |
| License: | Open Source |
| Edition: | Official |
| Input Labels: | [documents] |
| Output Labels: | [generation] |
| Language: | en |
| Size: | 1.6 GB |
References
https://huggingface.co/bigcode/starcoder2-3b