scenario-KD-PO-CDF-EN-FROM-EN-D2_data-en-cardiff_eng_only_beta-jason

This model is a fine-tuned version of haryoaw/scenario-TCR_data-en-cardiff_eng_only2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 26.5719
• Accuracy: 0.4056
• F1: 0.3992

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: 6666
• 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	1.72	100	21.2982	0.3690	0.3475
No log	3.45	200	21.1284	0.3792	0.3741
No log	5.17	300	21.4889	0.3893	0.3848
No log	6.9	400	22.1337	0.3995	0.3933
21.5838	8.62	500	22.7814	0.4092	0.3879
21.5838	10.34	600	22.4046	0.3946	0.3846
21.5838	12.07	700	23.0168	0.4017	0.4022
21.5838	13.79	800	22.9590	0.3946	0.3914
21.5838	15.52	900	23.6863	0.3964	0.3942
15.9629	17.24	1000	24.4297	0.3977	0.3898
15.9629	18.97	1100	24.2709	0.3981	0.3931
15.9629	20.69	1200	24.9858	0.4123	0.4129
15.9629	22.41	1300	25.4483	0.4083	0.4062
15.9629	24.14	1400	26.0959	0.3942	0.3925
11.4549	25.86	1500	26.1345	0.4056	0.4041
11.4549	27.59	1600	26.4943	0.3871	0.3812
11.4549	29.31	1700	26.5719	0.4056	0.3992

Framework versions

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