This model is a fine-tuned version of microsoft/mdeberta-v3-base on the tmnam20/VieGLUE/QNLI dataset. It achieves the following results on the evaluation set:

Loss: 0.2782
Accuracy: 0.8999

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: 32
eval_batch_size: 16
seed: 1
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3768	0.15	500	0.3291	0.8596
0.3506	0.31	1000	0.2961	0.8752
0.3417	0.46	1500	0.2917	0.8808
0.3319	0.61	2000	0.2742	0.8871
0.3126	0.76	2500	0.2686	0.8913
0.3073	0.92	3000	0.2639	0.8916
0.2867	1.07	3500	0.2557	0.8958
0.2313	1.22	4000	0.2937	0.8880
0.2364	1.37	4500	0.2585	0.8971
0.2533	1.53	5000	0.2545	0.8938
0.2333	1.68	5500	0.2629	0.8955
0.225	1.83	6000	0.2532	0.9002
0.2313	1.99	6500	0.2520	0.8988
0.1793	2.14	7000	0.2819	0.8953
0.1639	2.29	7500	0.2809	0.8964
0.1645	2.44	8000	0.2778	0.8990
0.1753	2.6	8500	0.2802	0.8988
0.1859	2.75	9000	0.2775	0.9001
0.1809	2.9	9500	0.2767	0.8988

Framework versions

Transformers 4.35.2
Pytorch 2.2.0.dev20231203+cu121
Datasets 2.15.0
Tokenizers 0.15.0

tmnam20/mdeberta-v3-base-qnli-1

mdeberta-v3-base-qnli-1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

tmnam20/
mdeberta-v3-base-qnli-1