AnnoMI-simple_speaker_role_id-bert-base-uncased-v1

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7799

• Accuracy: 0.8443

• Precision Macro: 0.8448

• Recall Macro: 0.8442

• F1 Macro: 0.8442

• Precision Weighted: 0.8447

• Recall Weighted: 0.8443

• F1 Weighted: 0.8443

• Report: precision recall f1-score support

       0       0.83      0.86      0.85       488
       1       0.86      0.83      0.84       482
  

  accuracy 0.84 970 macro avg 0.84 0.84 0.84 970

weighted avg 0.84 0.84 0.84 970

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Macro	Recall Macro	F1 Macro	Precision Weighted	Recall Weighted	F1 Weighted	Report
0.3142	2.0577	500	0.4281	0.8423	0.8438	0.8425	0.8421	0.8439	0.8423	0.8421	precision recall f1-score support

       0       0.87      0.81      0.84       488
       1       0.82      0.88      0.85       482

accuracy                           0.84       970

macro avg 0.84 0.84 0.84 970 weighted avg 0.84 0.84 0.84 970 | | 0.1986 | 4.1155 | 1000 | 0.5841 | 0.8144 | 0.8250 | 0.8150 | 0.8131 | 0.8253 | 0.8144 | 0.8130 | precision recall f1-score support

       0       0.89      0.73      0.80       488
       1       0.76      0.90      0.83       482

accuracy                           0.81       970

macro avg 0.82 0.81 0.81 970 weighted avg 0.83 0.81 0.81 970 | | 0.177 | 6.1732 | 1500 | 0.6025 | 0.8392 | 0.8392 | 0.8392 | 0.8392 | 0.8392 | 0.8392 | 0.8392 | precision recall f1-score support

       0       0.84      0.84      0.84       488
       1       0.84      0.84      0.84       482

accuracy                           0.84       970

macro avg 0.84 0.84 0.84 970 weighted avg 0.84 0.84 0.84 970 | | 0.1595 | 8.2309 | 2000 | 0.5967 | 0.8495 | 0.8495 | 0.8495 | 0.8495 | 0.8495 | 0.8495 | 0.8495 | precision recall f1-score support

       0       0.85      0.85      0.85       488
       1       0.85      0.84      0.85       482

accuracy                           0.85       970

macro avg 0.85 0.85 0.85 970 weighted avg 0.85 0.85 0.85 970 | | 0.1593 | 10.2887 | 2500 | 0.6821 | 0.8433 | 0.8433 | 0.8433 | 0.8433 | 0.8433 | 0.8433 | 0.8433 | precision recall f1-score support

       0       0.84      0.85      0.84       488
       1       0.85      0.84      0.84       482

accuracy                           0.84       970

macro avg 0.84 0.84 0.84 970 weighted avg 0.84 0.84 0.84 970 | | 0.148 | 12.3464 | 3000 | 0.6469 | 0.8423 | 0.8456 | 0.8420 | 0.8418 | 0.8454 | 0.8423 | 0.8418 | precision recall f1-score support

       0       0.81      0.89      0.85       488
       1       0.88      0.79      0.83       482

accuracy                           0.84       970

macro avg 0.85 0.84 0.84 970 weighted avg 0.85 0.84 0.84 970 | | 0.1446 | 14.4041 | 3500 | 0.7799 | 0.8443 | 0.8448 | 0.8442 | 0.8442 | 0.8447 | 0.8443 | 0.8443 | precision recall f1-score support

       0       0.83      0.86      0.85       488
       1       0.86      0.83      0.84       482

accuracy                           0.84       970

macro avg 0.84 0.84 0.84 970 weighted avg 0.84 0.84 0.84 970 |

Framework versions

• Transformers 4.57.6
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2