CeLLaTeNER-2class-bioformer8l-tapt-tokenizer-adapted-v1

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

CeLLaTeNER-bioformer8l-tapt-tokenizer-adapted-v1

This model is a fine-tuned version of Mardiyyah/CeLLaTe-bioformer-8L-tapt-adapted on the OTAR3088-2class NER dataset.

It achieves the following results on the evaluation set:

• Loss: 0.0699
• Precision: 0.6960
• Recall: 0.7332
• F1: 0.7141
• Accuracy: 0.9798

Model description

This is a TAPT + NER finetuned model, trained in multiple phase:

• Firstly, the base-model's tokenizer was adapted and extended to better represent biomedical domain terms and reduce fragmentation.
• Then, the adapted tokenizer was used in a continue-pretraining stage using the unlabelled task-specific corpus on Task-Adaptive finetuning(TAPT).
• Finally,the resulting tapt-model was fine-tuned on NER task using the labelled version of the data from step 2 above.

Label Schema

This model uses the following label space:

2-class

• CellLine
• Cell_Tissue

The label schema determines the classifier output dimension and the evaluation setup used for this model.

Base model

This model is based on:

• Architecture: Bioformer-8l
• Tokenizer: {tokenizer-adapted tokenizer}
• Pretraining checkpoint: bioformers/bioformer-8l

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 3407
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.48.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.2
• Tokenizers 0.21.0