scenario-NON-KD-SCR-COPY-CDF-CL-D2_data-cl-cardiff_cl_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: 6.2603
• Accuracy: 0.3657
• F1: 0.3633

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	1.0870	250	1.3365	0.3519	0.3479
0.9017	2.1739	500	1.9777	0.3565	0.3508
0.9017	3.2609	750	2.9438	0.3542	0.3298
0.3023	4.3478	1000	3.1702	0.3611	0.3489
0.3023	5.4348	1250	3.4689	0.3534	0.3522
0.1011	6.5217	1500	4.0537	0.3627	0.3608
0.1011	7.6087	1750	4.5352	0.3573	0.3504
0.0549	8.6957	2000	4.5030	0.3495	0.3449
0.0549	9.7826	2250	4.6084	0.3519	0.3479
0.0339	10.8696	2500	4.7223	0.3565	0.3505
0.0339	11.9565	2750	4.9936	0.3565	0.3518
0.0232	13.0435	3000	4.5828	0.3449	0.3352
0.0232	14.1304	3250	5.0265	0.3565	0.3543
0.0224	15.2174	3500	5.2273	0.3627	0.3580
0.0224	16.3043	3750	5.2708	0.3611	0.3516
0.0156	17.3913	4000	5.6845	0.3511	0.3469
0.0156	18.4783	4250	5.5643	0.3603	0.3537
0.0081	19.5652	4500	5.9288	0.3519	0.3372
0.0081	20.6522	4750	5.9406	0.3611	0.3564
0.0034	21.7391	5000	5.9909	0.3534	0.3519
0.0034	22.8261	5250	6.1283	0.3611	0.3562
0.0017	23.9130	5500	6.1721	0.3688	0.3668
0.0017	25.0	5750	6.2167	0.3596	0.3581
0.0019	26.0870	6000	6.2126	0.3627	0.3596
0.0019	27.1739	6250	6.2446	0.3634	0.3616
0.0014	28.2609	6500	6.2484	0.3650	0.3624
0.0014	29.3478	6750	6.2603	0.3657	0.3633

Framework versions

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