scenario-NON-KD-PR-COPY-CDF-EN-D2_data-en-cardiff_eng_only44

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: 5.1657
• Accuracy: 0.4577
• F1: 0.4543

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: 44
• 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	1.1068	0.4330	0.3826
No log	3.4483	200	1.4495	0.4533	0.4238
No log	5.1724	300	1.5295	0.4586	0.4497
No log	6.8966	400	2.0122	0.4537	0.4516
0.5768	8.6207	500	3.0885	0.4493	0.4417
0.5768	10.3448	600	3.3878	0.4541	0.4497
0.5768	12.0690	700	3.4115	0.4586	0.4564
0.5768	13.7931	800	3.8779	0.4590	0.4572
0.5768	15.5172	900	4.1514	0.4590	0.4579
0.0737	17.2414	1000	4.6699	0.4462	0.4281
0.0737	18.9655	1100	4.6724	0.4608	0.4612
0.0737	20.6897	1200	4.6790	0.4603	0.4562
0.0737	22.4138	1300	4.9305	0.4581	0.4564
0.0737	24.1379	1400	5.0621	0.4568	0.4503
0.0099	25.8621	1500	5.0787	0.4608	0.4574
0.0099	27.5862	1600	5.1428	0.4581	0.4549
0.0099	29.3103	1700	5.1657	0.4577	0.4543

Framework versions

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