scenario-KD-PO-CDF-EN-FROM-CL-D2_data-en-cardiff_eng_only44

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: 39.7248
• Accuracy: 0.4511
• F1: 0.4490

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: 44
• 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	25.4567	0.4422	0.4387
No log	3.45	200	25.1633	0.4793	0.4777
No log	5.17	300	30.1100	0.4356	0.4168
No log	6.9	400	30.5006	0.4815	0.4818
16.9844	8.62	500	32.4822	0.4493	0.4441
16.9844	10.34	600	35.6753	0.4541	0.4525
16.9844	12.07	700	37.9029	0.4669	0.4628
16.9844	13.79	800	40.5685	0.4462	0.4400
16.9844	15.52	900	39.8826	0.4489	0.4407
3.5242	17.24	1000	42.5352	0.4325	0.4211
3.5242	18.97	1100	36.9882	0.4696	0.4686
3.5242	20.69	1200	39.5527	0.4519	0.4478
3.5242	22.41	1300	40.9362	0.4295	0.4201
3.5242	24.14	1400	40.2929	0.4471	0.4446
1.4703	25.86	1500	39.7746	0.4568	0.4509
1.4703	27.59	1600	39.0260	0.4524	0.4503
1.4703	29.31	1700	39.7248	0.4511	0.4490

Framework versions

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