scenario-MDBT-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: 3.4982
• Accuracy: 0.5873
• F1: 0.5922

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: 64
• 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	0.9674	0.5220	0.5235
No log	2.07	120	0.9588	0.5847	0.5738
No log	3.1	180	1.1066	0.5789	0.5815
No log	4.14	240	1.2807	0.5547	0.5544
No log	5.17	300	1.4846	0.5899	0.5874
No log	6.21	360	1.8475	0.5564	0.5582
No log	7.24	420	2.1012	0.5679	0.5685
No log	8.28	480	2.2050	0.5855	0.5892
0.4223	9.31	540	2.3754	0.5807	0.5839
0.4223	10.34	600	2.4936	0.5798	0.5840
0.4223	11.38	660	2.7673	0.5631	0.5666
0.4223	12.41	720	2.5116	0.5798	0.5820
0.4223	13.45	780	2.7368	0.5622	0.5660
0.4223	14.48	840	2.7623	0.5763	0.5812
0.4223	15.52	900	2.8766	0.5833	0.5874
0.4223	16.55	960	2.8306	0.5838	0.5845
0.0534	17.59	1020	3.0452	0.5961	0.6000
0.0534	18.62	1080	3.1689	0.5825	0.5855
0.0534	19.66	1140	3.0359	0.5851	0.5893
0.0534	20.69	1200	3.4232	0.5754	0.5795
0.0534	21.72	1260	3.3304	0.5829	0.5876
0.0534	22.76	1320	3.3603	0.5864	0.5913
0.0534	23.79	1380	3.3638	0.5829	0.5876
0.0534	24.83	1440	3.3312	0.5930	0.5966
0.0065	25.86	1500	3.4156	0.5873	0.5919
0.0065	26.9	1560	3.4476	0.5877	0.5925
0.0065	27.93	1620	3.4884	0.5829	0.5878
0.0065	28.97	1680	3.4938	0.5842	0.5893
0.0065	30.0	1740	3.4982	0.5873	0.5922

Framework versions

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