mdeberta-v3-base-finetuned-temporal_focus-classification

This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7346
• Accuracy: 0.8510
• Accuracy Balanced: 0.8196
• F1 Macro: 0.8228
• F1 Weighted: 0.8508
• Precision Macro: 0.8262
• Recall Macro: 0.8196
• Precision Weighted: 0.8506
• Recall Weighted: 0.8510
• Mcc: 0.7409
• F1 Class0: 0.7548
• Precision Class0: 0.7695
• Recall Class0: 0.7406
• F1 Class1: 0.8277
• Precision Class1: 0.8249
• Recall Class1: 0.8304
• F1 Class2: 0.8861
• Precision Class2: 0.8843
• Recall Class2: 0.8879
• Roc Auc Ovr Macro: 0.9478

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: 8
• eval_batch_size: 8
• seed: 42
• 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_ratio: 0.06
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Accuracy Balanced	F1 Macro	F1 Weighted	Precision Macro	Recall Macro	Precision Weighted	Recall Weighted	Mcc	F1 Class0	Precision Class0	Recall Class0	F1 Class1	Precision Class1	Recall Class1	F1 Class2	Precision Class2	Recall Class2	Roc Auc Ovr Macro
0.4944	1.0	1570	0.4513	0.8354	0.7964	0.8027	0.8352	0.8145	0.7964	0.8390	0.8354	0.7175	0.7172	0.7774	0.6656	0.8210	0.7721	0.8767	0.8698	0.8939	0.8469	0.9445
0.4003	2.0	3140	0.5297	0.8424	0.7901	0.8012	0.8413	0.8178	0.7901	0.8427	0.8424	0.7259	0.6964	0.7719	0.6344	0.8228	0.7915	0.8568	0.8845	0.8900	0.8790	0.9480
0.31	3.0	4710	0.6586	0.8465	0.8062	0.8165	0.8444	0.8318	0.8062	0.8481	0.8465	0.7318	0.7528	0.7687	0.7375	0.8121	0.8824	0.7522	0.8845	0.8441	0.9289	0.9459
0.2306	4.0	6280	0.7346	0.8510	0.8196	0.8228	0.8508	0.8262	0.8196	0.8506	0.8510	0.7409	0.7548	0.7695	0.7406	0.8277	0.8249	0.8304	0.8861	0.8843	0.8879	0.9478

Framework versions

• Transformers 4.57.6
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2