biolinkbert-data-type-20261107

Fine-tuned model from the PsyNamic project (code, webapp): Data Type. This is the model deployed in the PsyNamic webapp after the revision (Nov 2026). It supersedes STRIDE-lab/biobert-data-type-20250221 (tag submission-v1, initial submission). All current models: PsyNamic collection.

Base Model

michiyasunaga/BioLinkBERT-base

Task

Data Type

Specifies whether data were collected longitudinally or cross-sectionally.

Labels (id2label)

ID Label Definition
0 Cross-sectional Outcomes are measured during one single time point (or more than one time point within ≀ 24 hours).
1 Longitudinal long Outcomes are measures during at least two different time points > 24 hours apart and the total follow-up time is β‰₯ 3 months.
2 Longitudinal short Outcomes are measures during at least two different time points > 24 hours apart but the total follow-up time is < 3 months.
3 Not applicable e.g., for systematic reviews or for studies pooling a variety of other studies together without providing information about data type.
4 Unknown Data type is unknown.

Prediction rule

Single-label: softmax over the labels, the label with the highest probability (argmax) is predicted.

Model selection

Five base models (BiomedBERT-abstract, BioBERT, BioLinkBERT, ClinicalBERT, SciBERT) were fine-tuned with identical hyperparameters. The checkpoint is the one with the lowest validation loss (early stopping), and the base model was selected on the validation set (weighted F1). The test set was used once, after selection.

Test Performance

Held-out test set (96 abstracts), argmax as in the webapp, 95% bootstrap CIs (1,000 BCa resamples).

Metric (test) Value [95% CI]
Weighted F1 0.712 [0.608–0.804]
Weighted precision 0.725 [0.600–0.806]
Weighted recall 0.729 [0.635–0.812]

Usage

import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification

repo = "STRIDE-lab/biolinkbert-data-type-20261107"
tokenizer = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSequenceClassification.from_pretrained(repo)
inputs = tokenizer("<title>. <abstract>", truncation=True, max_length=512, return_tensors="pt")
logits = model(**inputs).logits
pred = model.config.id2label[int(logits.argmax(-1))]

Training Parameters

{
  "data": "<private_dataset>",
  "task": "Data Type",
  "model": "biolinkbert",
  "cross_val": false,
  "batch_size": 8,
  "learning_rate": 5e-05,
  "weight_decay": 0.01,
  "lr_scheduler": "linear",
  "warmup_ratio": 0.1,
  "epochs": 30,
  "dropout": 0.1,
  "early_stopping_patience": 5,
  "gradient_clipping": 1.0,
  "max_length": 512,
  "threshold": 0.5,
  "is_multilabel": false
}

threshold (0.5) is only used to compute the validation metrics of multi-label tasks during training. It has no effect for this single-label model, which predicts by argmax.

Downloads last month
66
Safetensors
Model size
0.1B params
Tensor type
F32
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for STRIDE-lab/biolinkbert-data-type-20261107

Finetuned
(33)
this model

Collection including STRIDE-lab/biolinkbert-data-type-20261107