scenario-KD-PO-CDF-EN-FROM-CL-D2_data-en-cardiff_eng_only_alpha-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: 15.6158
• Accuracy: 0.4061
• F1: 0.3950

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: 2222
• 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	12.5428	0.3796	0.3146
No log	3.45	200	12.5657	0.3845	0.3414
No log	5.17	300	12.6557	0.3911	0.3631
No log	6.9	400	13.2594	0.4012	0.3807
13.0153	8.62	500	13.2352	0.3920	0.3836
13.0153	10.34	600	13.9522	0.3854	0.3638
13.0153	12.07	700	13.2917	0.3902	0.3891
13.0153	13.79	800	14.1281	0.3924	0.3931
13.0153	15.52	900	14.3501	0.4004	0.3889
9.1779	17.24	1000	14.2606	0.3955	0.3757
9.1779	18.97	1100	14.6113	0.4008	0.3953
9.1779	20.69	1200	14.9622	0.3973	0.3864
9.1779	22.41	1300	15.3755	0.3898	0.3652
9.1779	24.14	1400	15.1117	0.4087	0.4015
6.5007	25.86	1500	15.5084	0.4101	0.4075
6.5007	27.59	1600	15.5980	0.3955	0.3892
6.5007	29.31	1700	15.6158	0.4061	0.3950

Framework versions

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