scenario-KD-PR-CDF-EN-FROM-CL-D2_data-en-cardiff_eng_only_delta-jason

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 16.6296
• Accuracy: 0.3717
• F1: 0.3536

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: 7777
• 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	12.7327	0.3325	0.1812
No log	3.45	200	12.8292	0.3761	0.3217
No log	5.17	300	12.9466	0.3823	0.3524
No log	6.9	400	12.8228	0.3867	0.3832
13.1958	8.62	500	13.3388	0.3823	0.3761
13.1958	10.34	600	14.0438	0.3920	0.3814
13.1958	12.07	700	15.2932	0.3792	0.3468
13.1958	13.79	800	15.4804	0.3735	0.3295
13.1958	15.52	900	16.1070	0.3818	0.3438
8.893	17.24	1000	14.9589	0.3805	0.3482
8.893	18.97	1100	15.2843	0.3849	0.3704
8.893	20.69	1200	15.6003	0.3942	0.3873
8.893	22.41	1300	15.5817	0.4087	0.4044
8.893	24.14	1400	16.1571	0.3898	0.3802
6.538	25.86	1500	16.3557	0.3907	0.3763
6.538	27.59	1600	16.4529	0.3889	0.3753
6.538	29.31	1700	16.6296	0.3717	0.3536

Framework versions

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