Hashuz/AS_MentalQAU

results

This model is a fine-tuned version of UBC-NLP/MARBERTv2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4492
• F1-micro: 0.85
• F1-macro: 0.8331
• Jaccard: 0.7664

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 with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	F1-micro	F1-macro	Jaccard
No log	1.0	7	0.6044	0.7908	0.7281	0.6838
No log	2.0	14	0.5586	0.8010	0.7369	0.6947
No log	3.0	21	0.5412	0.8010	0.7369	0.6947
No log	4.0	28	0.5293	0.8010	0.7369	0.6947
No log	5.0	35	0.5149	0.8010	0.7369	0.6947
No log	6.0	42	0.5050	0.81	0.7527	0.7087
No log	7.0	49	0.4927	0.8170	0.7660	0.7212
No log	8.0	56	0.4856	0.82	0.7749	0.7243
No log	9.0	63	0.4770	0.8221	0.7768	0.7290
No log	10.0	70	0.4679	0.8279	0.7930	0.7336
No log	11.0	77	0.4615	0.835	0.8044	0.7461
No log	12.0	84	0.4545	0.8392	0.8128	0.7523
No log	13.0	91	0.4512	0.845	0.8254	0.7586
No log	14.0	98	0.4496	0.85	0.8331	0.7664
0.4157	15.0	105	0.4492	0.85	0.8331	0.7664

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.1