scenario-TCR_data-en-cardiff_eng_only

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: 3.5981
• Accuracy: 0.5798
• F1: 0.5830

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: 66
• 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	1.03	60	1.0880	0.5295	0.5251
No log	2.07	120	1.0869	0.5617	0.5392
No log	3.1	180	1.1333	0.5789	0.5818
No log	4.14	240	1.2897	0.5728	0.5743
No log	5.17	300	1.4495	0.5899	0.5944
No log	6.21	360	1.9107	0.5573	0.5582
No log	7.24	420	1.8983	0.5851	0.5883
No log	8.28	480	2.1481	0.5816	0.5838
0.4492	9.31	540	2.1906	0.5697	0.5681
0.4492	10.34	600	2.4558	0.5692	0.5658
0.4492	11.38	660	2.2698	0.5891	0.5917
0.4492	12.41	720	2.6192	0.5816	0.5832
0.4492	13.45	780	2.8040	0.5825	0.5866
0.4492	14.48	840	3.0573	0.5754	0.5790
0.4492	15.52	900	2.8448	0.5847	0.5872
0.4492	16.55	960	3.2238	0.5829	0.5874
0.0555	17.59	1020	3.2796	0.5811	0.5852
0.0555	18.62	1080	3.2371	0.5869	0.5878
0.0555	19.66	1140	3.4683	0.5802	0.5831
0.0555	20.69	1200	3.4679	0.5772	0.5793
0.0555	21.72	1260	3.4337	0.5877	0.5912
0.0555	22.76	1320	3.5059	0.5763	0.5792
0.0555	23.79	1380	3.6144	0.5807	0.5851
0.0555	24.83	1440	3.5076	0.5847	0.5874
0.0086	25.86	1500	3.5835	0.5842	0.5878
0.0086	26.9	1560	3.5517	0.5847	0.5872
0.0086	27.93	1620	3.6182	0.5825	0.5855
0.0086	28.97	1680	3.5885	0.5816	0.5847
0.0086	30.0	1740	3.5981	0.5798	0.5830

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.13.3