Description
This is a lightweight Text-to-SQL model fine-tuned by John Snow Labs, built specifically for working with medical and healthcare data.
How to use
from sparknlp.base import DocumentAssembler
from sparknlp_jsl.annotator import MedicalLLM
from pyspark.ml import Pipeline
document_assembler = DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("document")
medical_llm = MedicalLLM.pretrained("jsl_meds_text2sql_1b_q16_v1", "en", "clinical/models")\
.setInputCols("document")\
.setOutputCol("completions")\
.setBatchSize(1)\
.setNPredict(100)\
.setUseChatTemplate(True)\
.setTemperature(0)
pipeline = Pipeline(stages=[
document_assembler,
medical_llm
])
medm_prompt = """### Instruction:
### Instruction:
Table: CancerPatients
- patient_id (INT)
- name (VARCHAR)
- age (INT)
- gender (VARCHAR)
- cancer_type (VARCHAR)
- diagnosis_date (DATE)
List the names of patients diagnosed with breast cancer.
### Response:
"""
data = spark.createDataFrame([[medm_prompt]]).toDF("text")
model = pipeline.fit(data)
result = model.transform(data)
result.select("completions").show(truncate=False)
from johnsnowlabs import nlp, medical
document_assembler = nlp.DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("document")
medical_llm = medical.MedicalLLM.pretrained(jsl_meds_text2sql_1b_q16_v1, "en", "clinical/models")\
.setInputCols("document")\
.setOutputCol("completions")\
.setBatchSize(1)\
.setNPredict(100)\
.setUseChatTemplate(True)\
.setTemperature(0)
pipeline = nlp.Pipeline(stages=[
document_assembler,
medical_llm
])
medm_prompt = """### Instruction:
### Instruction:
Table: CancerPatients
- patient_id (INT)
- name (VARCHAR)
- age (INT)
- gender (VARCHAR)
- cancer_type (VARCHAR)
- diagnosis_date (DATE)
List the names of patients diagnosed with breast cancer.
### Response:
"""
data = spark.createDataFrame([[medm_prompt]]).toDF("text")
model = pipeline.fit(data)
result = model.transform(data)
result.select("completions").show(truncate=False)
val documentAssembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val medicalLLM = MedicalLLM.pretrained("jsl_meds_text2sql_1b_q16_v1", "en", "clinical/models")
.setInputCols("document")
.setOutputCol("completions")
.setBatchSize(1)
.setNPredict(100)
.setUseChatTemplate(true)
.setTemperature(0)
val pipeline = new Pipeline().setStages(Array(
documentAssembler,
medicalLLM
))
val medmPrompt = """### Instruction:
### Instruction:
Table: CancerPatients
- patient_id (INT)
- name (VARCHAR)
- age (INT)
- gender (VARCHAR)
- cancer_type (VARCHAR)
- diagnosis_date (DATE)
List the names of patients diagnosed with breast cancer.
### Response:
"""
val data = Seq(medmPrompt).toDF("text")
val model = pipeline.fit(data)
val result = model.transform(data)
result.select("completions").show(false)
Results
SELECT name FROM CancerPatients WHERE cancer_type = 'breast cancer'
Model Information
| Model Name: | jsl_meds_text2sql_1b_q16_v1 |
| Compatibility: | Healthcare NLP 6.1.0+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
| Size: | 1.6 GB |