scenario-NON-KD-PO-COPY-CDF-EN-D2_data-en-cardiff_eng_only44

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

• Loss: 5.0452
• Accuracy: 0.4590
• F1: 0.4571

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	1.0902	0.4237	0.3991
No log	3.45	200	1.6849	0.4449	0.4221
No log	5.17	300	2.0491	0.4321	0.4048
No log	6.9	400	2.6699	0.4405	0.4278
0.5568	8.62	500	2.9052	0.4515	0.4471
0.5568	10.34	600	3.2156	0.4515	0.4466
0.5568	12.07	700	3.0671	0.4502	0.4497
0.5568	13.79	800	3.7902	0.4528	0.4501
0.5568	15.52	900	4.0797	0.4577	0.4561
0.0772	17.24	1000	4.3812	0.4577	0.4508
0.0772	18.97	1100	4.6449	0.4555	0.4463
0.0772	20.69	1200	4.7051	0.4440	0.4327
0.0772	22.41	1300	4.9485	0.4458	0.4363
0.0772	24.14	1400	4.9522	0.4528	0.4471
0.0147	25.86	1500	4.9600	0.4572	0.4564
0.0147	27.59	1600	5.0399	0.4533	0.4503
0.0147	29.31	1700	5.0452	0.4590	0.4571

Framework versions

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