scenario-NON-KD-PR-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: 5.7984
• Accuracy: 0.4336
• F1: 0.4331

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	1.2895	0.4267	0.4234
0.9107	2.1739	500	1.4607	0.4252	0.4153
0.9107	3.2609	750	1.7315	0.4460	0.4429
0.5589	4.3478	1000	1.9288	0.4375	0.4350
0.5589	5.4348	1250	2.3346	0.4390	0.4355
0.2715	6.5217	1500	2.4616	0.4491	0.4484
0.2715	7.6087	1750	3.6130	0.4321	0.4302
0.1449	8.6957	2000	3.1468	0.4498	0.4496
0.1449	9.7826	2250	3.5067	0.4522	0.4521
0.0935	10.8696	2500	3.7250	0.4414	0.4385
0.0935	11.9565	2750	4.2294	0.4275	0.4257
0.0612	13.0435	3000	4.3569	0.4198	0.4164
0.0612	14.1304	3250	4.9762	0.4113	0.3998
0.0488	15.2174	3500	5.2506	0.4367	0.4233
0.0488	16.3043	3750	4.9138	0.4329	0.4273
0.0283	17.3913	4000	4.7608	0.4267	0.4238
0.0283	18.4783	4250	5.0986	0.4429	0.4412
0.0235	19.5652	4500	5.0181	0.4475	0.4472
0.0235	20.6522	4750	5.4038	0.4437	0.4433
0.0167	21.7391	5000	5.4525	0.4383	0.4372
0.0167	22.8261	5250	5.7268	0.4398	0.4394
0.0084	23.9130	5500	6.0640	0.4329	0.4303
0.0084	25.0	5750	5.9652	0.4290	0.4264
0.0118	26.0870	6000	5.8877	0.4367	0.4352
0.0118	27.1739	6250	5.8917	0.4267	0.4236
0.0081	28.2609	6500	5.9397	0.4321	0.4292
0.0081	29.3478	6750	5.7984	0.4336	0.4331

Framework versions

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