scenario-KD-PO-CDF-EN-FROM-CL-D2_data-en-cardiff_eng_only_gamma-jason

This model is a fine-tuned version of haryoaw/scenario-TCR_data-cl-cardiff_cl_only2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 16.7514
• Accuracy: 0.3977
• F1: 0.3868

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: 88458
• 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	12.6439	0.3228	0.2712
No log	3.45	200	12.7249	0.3937	0.3882
No log	5.17	300	12.8006	0.3902	0.3809
No log	6.9	400	14.5097	0.3686	0.2952
12.7089	8.62	500	13.4001	0.4017	0.3928
12.7089	10.34	600	14.3419	0.4198	0.4042
12.7089	12.07	700	14.6078	0.3907	0.3741
12.7089	13.79	800	13.9429	0.3981	0.3942
12.7089	15.52	900	14.5756	0.3999	0.4000
8.5226	17.24	1000	14.8560	0.4012	0.3983
8.5226	18.97	1100	15.4018	0.4061	0.3972
8.5226	20.69	1200	15.6938	0.4127	0.4041
8.5226	22.41	1300	15.9840	0.4004	0.3879
8.5226	24.14	1400	15.9819	0.3977	0.3829
6.1231	25.86	1500	16.0163	0.4052	0.3938
6.1231	27.59	1600	16.2095	0.3968	0.3883
6.1231	29.31	1700	16.7514	0.3977	0.3868

Framework versions

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