tsilva/clinical-field-mapper-causal_lm

Model Card for tsilva/clinical-field-mapper-causal_lm

This model is a fine-tuned version of distilbert/distilgpt2 on the `tsilva/clinical-field-mappings` dataset. Its purpose is to normalize healthcare database column names to a standardized set of target column names.

Task

This is a causal language model designed to map free-text field names to standardized schema terms.

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.frompretrained("tsilva/clinical-field-mapper-causallm") model = AutoModelForCausalLM.frompretrained("tsilva/clinical-field-mapper-causallm")

def predict(inputtext): inputs = tokenizer(inputtext + "|", returntensors="pt") outputs = model.generate(**inputs, maxnewtokens=50) print(tokenizer.decode(outputs[0], skipspecial_tokens=True))

predict('cardi@')

Evaluation Results

• —train accuracy: 98.24%
• —validation accuracy: 89.84%
• —test accuracy: 89.35%

Training Details

• —Seed: 42
• —Epochs scheduled: 50
• —Epochs completed: 14
• —Early stopping triggered: Yes
• —Final training loss: 1.3344
• —Final evaluation loss: 1.1981
• —Optimizer: adamwbnb8bit
• —Learning rate: 0.0005
• —Batch size: 512
• —Precision: fp16
• —DeepSpeed enabled: True
• —Gradient accumulation steps: 1

License

Specify your license here (e.g., Apache 2.0, MIT, etc.)

Limitations and Bias

• —Model was trained on a specific clinical mapping dataset.
• —Performance may vary on out-of-distribution column names.
• —Ensure you validate model outputs in production environments.