Text Generation
PEFT
English
gemma-3
lora
clinical-nlp
healthcare
hipaa-safe-harbor
open-science
pocketgull
Instructions to use philgear/gemma-3-phenopackets-genomics with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use philgear/gemma-3-phenopackets-genomics with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("google/gemma-3-12b-it") model = PeftModel.from_pretrained(base_model, "philgear/gemma-3-phenopackets-genomics") - Notebooks
- Google Colab
- Kaggle
ποΈ PocketGull Gemma 3 GA4GH Phenopackets v2 Translator
Organization: PocketGull LLC (Oregon Registry: 258869891)
Informatics Lead: Phillip Gear (CMS NPI: 1487569752 | ORCID: 0009-0008-1372-5381)
Base Foundation Model: google/gemma-3-12b-it
Discipline: Rare Disease Genomics & HPO Ontology
Open Science Provenance: Zenodo DOI 10.5281/zenodo.20647514
π Overview
Translates free-text clinical notes into compliant GA4GH Phenopackets Schema v2.0 JSON with HPO/LOINC concepts.
This LoRA adapter was fine-tuned using Direct Preference Optimization (DPO) on domain-specific clinical datasets conforming strictly to HIPAA Β§164.514 Safe Harbor de-identification standards.
π Quickstart Inference (Transformers & PEFT)
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel
base_model_id = "google/gemma-3-12b-it"
adapter_id = "philgear/gemma-3-phenopackets-genomics"
tokenizer = AutoTokenizer.from_pretrained(base_model_id)
base_model = AutoModelForCausalLM.from_pretrained(
base_model_id,
torch_dtype=torch.bfloat16,
device_map="auto"
)
model = PeftModel.from_pretrained(base_model, adapter_id)
prompt = "Translate clinical phenotype into GA4GH Phenopackets v2.0 JSON: Proband with progressive cerebellar ataxia (HP:0001251), dysarthria (HP:0001260), and pathogenic ATXN3 repeat."
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
with torch.no_grad():
outputs = model.generate(**inputs, max_new_tokens=256, temperature=0.2)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
π HIPAA & Regulatory Compliance
- Zero-PHI Retention: Designed for local edge computation and private Google Cloud Vertex AI deployment.
- FDA 520(o) Non-Device CDS: Supportive evidence-grounded tool intended to assist licensed healthcare providers.
π Citation
@software{pocketgull_clinical_2026,
author = {Gear, Phillip},
title = {Pocket-Gull: Living Medical Intelligence Engine & Open Clinical Science Suite},
publisher = {Zenodo},
version = {1.25.0},
year = {2026},
doi = {10.5281/zenodo.20647514},
url = {https://pocketgull.app}
}
- Downloads last month
- 22