metadata

library_name: transformers
license: mit
base_model: microsoft/mdeberta-v3-base
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
model-index:
  - name: CS221-mdeberta-v3-base-randomdrop
    results: []

CS221-mdeberta-v3-base-randomdrop

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: 0.5440
F1: 0.6741
Roc Auc: 0.7756
Accuracy: 0.4071

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.5661	1.0	99	0.5434	0.0	0.5	0.1425
0.5054	2.0	198	0.4744	0.4852	0.6560	0.2621
0.4409	3.0	297	0.4436	0.5766	0.7104	0.3308
0.3975	4.0	396	0.4284	0.6071	0.7316	0.3588
0.2827	5.0	495	0.4228	0.6095	0.7296	0.3562
0.2831	6.0	594	0.4540	0.6467	0.7642	0.3715
0.1846	7.0	693	0.4519	0.6325	0.7459	0.3893
0.1752	8.0	792	0.4538	0.6426	0.7535	0.3740
0.1547	9.0	891	0.4799	0.6541	0.7642	0.3791
0.1046	10.0	990	0.4793	0.6667	0.7687	0.4020
0.1052	11.0	1089	0.5001	0.6593	0.7658	0.4046
0.0843	12.0	1188	0.5069	0.6647	0.7705	0.3893
0.0653	13.0	1287	0.5275	0.6681	0.7669	0.4097
0.0575	14.0	1386	0.5455	0.6617	0.7632	0.3944
0.0503	15.0	1485	0.5440	0.6741	0.7756	0.4071
0.0499	16.0	1584	0.5555	0.6653	0.7660	0.4097
0.0431	17.0	1683	0.5557	0.6660	0.7675	0.4020
0.0422	18.0	1782	0.5599	0.6632	0.7664	0.3944

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0