Flamenco43 / bert_800_abstracts_NER

bert_800_abstracts_NER

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4059
• Precision: 0.5876
• Recall: 0.6756
• F1: 0.6285
• Accuracy: 0.8788
• Per Tag Metrics: {'I-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9902903892834667}, 'I-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9926728400611345}, 'B-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9952800503461297}, 'I-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9890766879439}, 'B-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9784230872965927}, 'I-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9925829362582037}, 'I-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9902004854805359}, 'I-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9993706733794839}, 'O': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9291108513890137}, 'B-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9714555425694507}, 'B-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.992403128652342}, 'B-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9893014474512272}, 'I-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9768048188438371}, 'B-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9912793311157062}, 'B-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9793670772273667}}

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Framework versions

• Transformers 4.41.2
• Pytorch 2.2.1+cu118
• Datasets 2.19.1
• Tokenizers 0.19.1