scenario-NON-KD-PR-COPY-CDF-EN-D2_data-en-cardiff_eng_only55

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: 4.8923
• Accuracy: 0.4643
• F1: 0.4592

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 55
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.7241	100	1.2063	0.3911	0.3238
No log	3.4483	200	1.3163	0.4528	0.4320
No log	5.1724	300	1.6209	0.4691	0.4685
No log	6.8966	400	1.7389	0.4660	0.4638
0.6047	8.6207	500	2.6389	0.4616	0.4489
0.6047	10.3448	600	3.1051	0.4537	0.4455
0.6047	12.0690	700	3.2470	0.4581	0.4508
0.6047	13.7931	800	3.8151	0.4519	0.4466
0.6047	15.5172	900	3.9579	0.4625	0.4590
0.0877	17.2414	1000	3.9385	0.4718	0.4706
0.0877	18.9655	1100	4.6107	0.4568	0.4523
0.0877	20.6897	1200	4.7369	0.4568	0.4493
0.0877	22.4138	1300	4.7006	0.4700	0.4659
0.0877	24.1379	1400	4.7994	0.4674	0.4630
0.0116	25.8621	1500	4.8787	0.4612	0.4559
0.0116	27.5862	1600	4.8899	0.4660	0.4611
0.0116	29.3103	1700	4.8923	0.4643	0.4592

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.19.1