mdeberta-v3-base-finetuned-green-classification-full

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: 0.2485
• Accuracy: 0.9517
• Accuracy Balanced: 0.9463
• F1 Macro: 0.9498
• F1 Weighted: 0.9515
• F1 Micro: 0.9517
• F1 Positive: 0.9596
• Precision Macro: 0.9544
• Precision Weighted: 0.9523
• Recall Macro: 0.9463
• Recall Weighted: 0.9517
• Mcc: 0.9006
• Roc Auc: 0.9897
• Pr Auc: 0.9926

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	F1 Micro	F1 Positive	Precision Macro	Precision Weighted	Recall Macro	Recall Weighted	Mcc	Roc Auc	Pr Auc
0.2525	1.0	2102	0.2504	0.9409	0.9350	0.9385	0.9406	0.9409	0.9505	0.9432	0.9414	0.9350	0.9409	0.8781	0.9853	0.9896
0.1641	2.0	4204	0.2428	0.9489	0.9495	0.9476	0.9490	0.9489	0.9560	0.9459	0.9494	0.9495	0.9489	0.8955	0.9869	0.9912
0.0946	3.0	6306	0.2533	0.9536	0.9492	0.9519	0.9535	0.9536	0.9610	0.9552	0.9539	0.9492	0.9536	0.9044	0.9895	0.9925
0.0529	4.0	8408	0.2905	0.9531	0.9495	0.9514	0.9530	0.9531	0.9604	0.9536	0.9531	0.9495	0.9531	0.9032	0.9895	0.9926

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1