Text Classification
Transformers
Safetensors
English
Chinese
qwen3_5
image-text-to-text
decision-model
system-one
biomedical
clinical-trials
pharmacovigilance
regulatory
genomics
llm-as-a-judge
calibration
jev
typed-decisions
calibrated-probabilities
Eval Results (legacy)
Instructions to use lighteternal/biodecision-v2-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lighteternal/biodecision-v2-4b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="lighteternal/biodecision-v2-4b")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("lighteternal/biodecision-v2-4b") model = AutoModelForMultimodalLM.from_pretrained("lighteternal/biodecision-v2-4b", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download temperatures.json from lighteternal/biodecision-v2-4b: direct link, hf CLI and curl.
- Browser
- Download file 2.63 kB
-
https://huggingface.co/lighteternal/biodecision-v2-4b/resolve/main/temperatures.json
- Command line
-
hf download hf://lighteternal/biodecision-v2-4b/temperatures.json
-
curl -L -o temperatures.json https://huggingface.co/lighteternal/biodecision-v2-4b/resolve/main/temperatures.json
2.63 kB
| { | |
| "global": 1.0613682609912578, | |
| "families": { | |
| "exams_en": 1.0203639538117337, | |
| "evidence_qa": 1.06402833182081, | |
| "claim_nli": 1.3632992277211728, | |
| "trial_outcome": 1.0076728598588514, | |
| "relations": 1.0947156498391535, | |
| "pharmacovigilance": 0.9967196838401151, | |
| "drug_reviews": 0.973623308078599, | |
| "med_info": 0.7586463380696176, | |
| "genomics": 0.973623308078599, | |
| "general": 0.9967196838401151, | |
| "judge": 1.240054182283984, | |
| "standards_coding": 1.2083282534652304, | |
| "trial_design": 1.135853947067939, | |
| "knowledge_graph": 1.0865461300642343, | |
| "diagnosis": 1.1234723611436637, | |
| "exams_zh": 1.0102110786494278, | |
| "regulatory": 1.0497659764565654 | |
| }, | |
| "family_of_source": { | |
| "ade_corpus": "pharmacovigilance", | |
| "bc5cdr": "relations", | |
| "bionli": "claim_nli", | |
| "chia": "trial_design", | |
| "civic": "genomics", | |
| "clinvar": "genomics", | |
| "cmb": "exams_zh", | |
| "cmexam": "exams_zh", | |
| "cosmos": "standards_coding", | |
| "ct_open_train": "trial_outcome", | |
| "ctgov_usdm": "trial_design", | |
| "ddi_corpus": "relations", | |
| "ddxplus": "diagnosis", | |
| "drug_reviews_condition": "drug_reviews", | |
| "drug_reviews_rating": "drug_reviews", | |
| "drugprot": "relations", | |
| "evidence_inference": "evidence_qa", | |
| "fda_enforcement": "regulatory", | |
| "fda_warning_letters": "regulatory", | |
| "gad": "genomics", | |
| "grounding_synth": "judge", | |
| "headqa": "exams_en", | |
| "judge_synth": "judge", | |
| "maude": "pharmacovigilance", | |
| "mcq_noul": "judge", | |
| "medconceptsqa": "standards_coding", | |
| "medhallu_artificial": "judge", | |
| "medical_abstracts": "evidence_qa", | |
| "medical_question_pairs": "claim_nli", | |
| "medinfo_qtype": "med_info", | |
| "mediqa_rqe": "claim_nli", | |
| "medmcqa": "exams_en", | |
| "medqa": "exams_en", | |
| "medquad": "med_info", | |
| "nli4ct": "claim_nli", | |
| "onsides": "pharmacovigilance", | |
| "opentargets": "genomics", | |
| "phee": "pharmacovigilance", | |
| "primekg": "knowledge_graph", | |
| "pubhealth": "claim_nli", | |
| "pubmed_rct": "evidence_qa", | |
| "pubmedqa_artificial": "evidence_qa", | |
| "pubmedqa_labeled": "evidence_qa", | |
| "scifact": "claim_nli", | |
| "sciq": "exams_en", | |
| "spl_sections": "regulatory", | |
| "tev1_general": "general", | |
| "trec_ct": "trial_matching", | |
| "trialbench_adverse_event": "trial_outcome", | |
| "trialbench_mortality_rate": "trial_outcome", | |
| "trialbench_patient_dropout": "trial_outcome", | |
| "trialbench_trial_approval": "trial_outcome", | |
| "trialbench_trial_duration": "trial_outcome", | |
| "trialbench_trial_failure_reason": "trial_outcome", | |
| "trialgpt_criteria": "trial_matching", | |
| "trialpanorama": "trial_design", | |
| "ultramedical_medqa_evol": "exams_en", | |
| "ultramedical_textbookqa": "exams_en", | |
| "vaers": "pharmacovigilance" | |
| } | |
| } |