metadata

library_name: transformers
license: mit
base_model: microsoft/deberta-base
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
model-index:
  - name: CS221-deberta-base-finetuned-semeval-aug
    results: []

CS221-deberta-base-finetuned-semeval-aug

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

Loss: 0.3475
F1: 0.8781
Roc Auc: 0.9070
Accuracy: 0.7651

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
0.4197	1.0	277	0.3731	0.6624	0.7504	0.4435
0.2694	2.0	554	0.3172	0.7734	0.8309	0.5411
0.1796	3.0	831	0.2815	0.7769	0.8256	0.5890
0.1281	4.0	1108	0.2802	0.8120	0.8543	0.6305
0.0754	5.0	1385	0.2998	0.8177	0.8565	0.6495
0.067	6.0	1662	0.2926	0.8367	0.8755	0.6838
0.0303	7.0	1939	0.2977	0.8409	0.8750	0.7010
0.009	8.0	2216	0.3252	0.8474	0.8777	0.7091
0.0114	9.0	2493	0.3181	0.8539	0.8899	0.7281
0.006	10.0	2770	0.3390	0.8581	0.8890	0.7344
0.0023	11.0	3047	0.3407	0.8646	0.8934	0.7353
0.0022	12.0	3324	0.3453	0.8674	0.8991	0.7525
0.0031	13.0	3601	0.3488	0.8708	0.9021	0.7507
0.0013	14.0	3878	0.3440	0.8736	0.9044	0.7579
0.0009	15.0	4155	0.3475	0.8781	0.9070	0.7651
0.0026	16.0	4432	0.3455	0.8767	0.9057	0.7651
0.0008	17.0	4709	0.3504	0.8755	0.9053	0.7615
0.0009	18.0	4986	0.3549	0.8742	0.9043	0.7588

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0