scenario-KD-PR-CDF-EN-FROM-CL-D2_data-en-cardiff_eng_only66

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

• Loss: 1.3328
• Accuracy: 0.4881
• F1: 0.4886

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: 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	1.72	100	1.3021	0.4947	0.4907
No log	3.45	200	1.3343	0.4713	0.4544
No log	5.17	300	1.3753	0.4899	0.4851
No log	6.9	400	1.3847	0.4572	0.4442
1.1115	8.62	500	1.3543	0.4757	0.4743
1.1115	10.34	600	1.3693	0.4691	0.4612
1.1115	12.07	700	1.3906	0.4687	0.4600
1.1115	13.79	800	1.3669	0.4700	0.4635
1.1115	15.52	900	1.3459	0.4899	0.4892
0.9572	17.24	1000	1.3562	0.4912	0.4872
0.9572	18.97	1100	1.3492	0.4846	0.4842
0.9572	20.69	1200	1.3918	0.4660	0.4504
0.9572	22.41	1300	1.3279	0.4960	0.4932
0.9572	24.14	1400	1.3440	0.4832	0.4832
0.936	25.86	1500	1.3460	0.4863	0.4863
0.936	27.59	1600	1.3371	0.4907	0.4909
0.936	29.31	1700	1.3328	0.4881	0.4886

Framework versions

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