scenario-TCR-XLMR-XCOPA_data-xcopa_all

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

• Loss: 0.6932
• Accuracy: 0.4558
• F1: 0.4185

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: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.38	5	0.6932	0.49	0.4715
No log	0.77	10	0.6932	0.4925	0.4634
No log	1.15	15	0.6931	0.5033	0.4669
No log	1.54	20	0.6932	0.4683	0.4491
No log	1.92	25	0.6932	0.4883	0.4689
No log	2.31	30	0.6932	0.4808	0.4716
No log	2.69	35	0.6931	0.5058	0.4785
No log	3.08	40	0.6932	0.4867	0.4615
No log	3.46	45	0.6932	0.4533	0.4090
No log	3.85	50	0.6932	0.4592	0.4190
No log	4.23	55	0.6931	0.5167	0.4858
No log	4.62	60	0.6931	0.5133	0.4832
No log	5.0	65	0.6931	0.525	0.5128
No log	5.38	70	0.6931	0.5158	0.5072
No log	5.77	75	0.6931	0.5275	0.5158
No log	6.15	80	0.6931	0.5333	0.5302
No log	6.54	85	0.6931	0.5075	0.4883
No log	6.92	90	0.6931	0.5042	0.4893
No log	7.31	95	0.6930	0.5042	0.4979
No log	7.69	100	0.6932	0.4883	0.4623
No log	8.08	105	0.6932	0.4617	0.4159
No log	8.46	110	0.6932	0.4425	0.4085
No log	8.85	115	0.6931	0.4992	0.4705
No log	9.23	120	0.6931	0.4975	0.4668
No log	9.62	125	0.6931	0.4867	0.4510
No log	10.0	130	0.6931	0.4817	0.4456
No log	10.38	135	0.6931	0.4808	0.4442
No log	10.77	140	0.6932	0.49	0.4565
No log	11.15	145	0.6932	0.4908	0.4579
No log	11.54	150	0.6931	0.5067	0.4761
No log	11.92	155	0.6931	0.5083	0.4887
No log	12.31	160	0.6931	0.5217	0.4974
No log	12.69	165	0.6932	0.4967	0.4499
No log	13.08	170	0.6931	0.5158	0.4872
No log	13.46	175	0.6931	0.5225	0.5065
No log	13.85	180	0.6931	0.5475	0.5323
No log	14.23	185	0.6931	0.5383	0.5224
No log	14.62	190	0.6932	0.4925	0.4558
No log	15.0	195	0.6932	0.4783	0.4340
No log	15.38	200	0.6931	0.5267	0.5120
No log	15.77	205	0.6931	0.5158	0.4890
No log	16.15	210	0.6931	0.53	0.5179
No log	16.54	215	0.6932	0.48	0.4497
No log	16.92	220	0.6932	0.4558	0.4133
No log	17.31	225	0.6932	0.4742	0.4290
No log	17.69	230	0.6931	0.4967	0.4838
No log	18.08	235	0.6931	0.5325	0.5126
No log	18.46	240	0.6931	0.5242	0.5090
No log	18.85	245	0.6931	0.5242	0.5013
No log	19.23	250	0.6932	0.4683	0.4273
No log	19.62	255	0.6931	0.5158	0.4890
No log	20.0	260	0.6932	0.4533	0.4225
No log	20.38	265	0.6932	0.4658	0.4282
No log	20.77	270	0.6932	0.4667	0.4326
No log	21.15	275	0.6932	0.4725	0.4403
No log	21.54	280	0.6932	0.4667	0.4306
No log	21.92	285	0.6932	0.4658	0.4272
No log	22.31	290	0.6932	0.4625	0.4256
No log	22.69	295	0.6932	0.4617	0.4232
No log	23.08	300	0.6932	0.465	0.4247
No log	23.46	305	0.6932	0.455	0.4223
No log	23.85	310	0.6931	0.4558	0.4226
No log	24.23	315	0.6931	0.4875	0.4638
No log	24.62	320	0.6931	0.4642	0.4315
No log	25.0	325	0.6931	0.4508	0.4153
No log	25.38	330	0.6932	0.4558	0.4185

Framework versions

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