scenario-KD-PO-CDF-EN-FROM-EN-D2_data-en-cardiff_eng_only_delta-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: 26.0242
• Accuracy: 0.3907
• F1: 0.3859

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: 7777
• 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	21.4135	0.3382	0.1963
No log	3.45	200	21.5272	0.3845	0.3506
No log	5.17	300	22.3621	0.3748	0.3274
No log	6.9	400	21.4286	0.3959	0.3950
21.7005	8.62	500	22.3157	0.3937	0.3766
21.7005	10.34	600	22.4956	0.4065	0.4025
21.7005	12.07	700	22.5078	0.4070	0.4061
21.7005	13.79	800	23.7347	0.4017	0.3816
21.7005	15.52	900	23.6096	0.3964	0.3828
15.6576	17.24	1000	24.2553	0.3836	0.3594
15.6576	18.97	1100	24.5136	0.3876	0.3805
15.6576	20.69	1200	24.4989	0.3986	0.3963
15.6576	22.41	1300	24.8175	0.4149	0.4117
15.6576	24.14	1400	25.9234	0.4008	0.3884
11.4297	25.86	1500	25.3551	0.4026	0.3979
11.4297	27.59	1600	25.9706	0.3995	0.3952
11.4297	29.31	1700	26.0242	0.3907	0.3859

Framework versions

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