scenario-TCR-XLMV-XCOPA-5_data-xcopa_all

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

• Loss: 0.6931
• Accuracy: 0.505
• F1: 0.4396

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: 214
• 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	0.6937	0.5158	0.4808
No log	0.77	10	0.6929	0.4917	0.4777
No log	1.15	15	0.6933	0.5025	0.4919
No log	1.54	20	0.6945	0.5125	0.4734
No log	1.92	25	0.6931	0.4767	0.4442
No log	2.31	30	0.6931	0.5192	0.4898
No log	2.69	35	0.6931	0.5042	0.4644
No log	3.08	40	0.6931	0.5183	0.5085
No log	3.46	45	0.6931	0.5083	0.4896
No log	3.85	50	0.6931	0.5158	0.5004
No log	4.23	55	0.6931	0.5292	0.5031
No log	4.62	60	0.6931	0.5233	0.5077
No log	5.0	65	0.6931	0.5292	0.5083
No log	5.38	70	0.6931	0.5	0.4643
No log	5.77	75	0.6931	0.4983	0.4837
No log	6.15	80	0.6932	0.4708	0.4385
No log	6.54	85	0.6932	0.475	0.4375
No log	6.92	90	0.6932	0.4717	0.4379
No log	7.31	95	0.6932	0.4925	0.4518
No log	7.69	100	0.6932	0.4758	0.4448
No log	8.08	105	0.6932	0.4908	0.4617
No log	8.46	110	0.6931	0.5017	0.4613
No log	8.85	115	0.6931	0.4875	0.4591
No log	9.23	120	0.6931	0.4783	0.4371
No log	9.62	125	0.6932	0.4533	0.4164
No log	10.0	130	0.6932	0.475	0.4335
No log	10.38	135	0.6931	0.5242	0.4870
No log	10.77	140	0.6931	0.5258	0.4969
No log	11.15	145	0.6931	0.4808	0.4452
No log	11.54	150	0.6932	0.4408	0.3993
No log	11.92	155	0.6932	0.4375	0.3968
No log	12.31	160	0.6932	0.4325	0.3925
No log	12.69	165	0.6932	0.435	0.4084
No log	13.08	170	0.6932	0.4825	0.4557
No log	13.46	175	0.6931	0.4892	0.4541
No log	13.85	180	0.6931	0.5	0.4828
No log	14.23	185	0.6931	0.5325	0.5109
No log	14.62	190	0.6931	0.5367	0.5071
No log	15.0	195	0.6932	0.47	0.4291
No log	15.38	200	0.6932	0.4483	0.4193
No log	15.77	205	0.6932	0.4325	0.4052
No log	16.15	210	0.6932	0.47	0.4411
No log	16.54	215	0.6932	0.47	0.4441
No log	16.92	220	0.6932	0.4567	0.4210
No log	17.31	225	0.6932	0.4408	0.4067
No log	17.69	230	0.6932	0.44	0.4032
No log	18.08	235	0.6932	0.455	0.4233
No log	18.46	240	0.6932	0.4558	0.4297
No log	18.85	245	0.6932	0.4542	0.4239
No log	19.23	250	0.6932	0.4783	0.4431
No log	19.62	255	0.6931	0.4908	0.4664
No log	20.0	260	0.6932	0.4825	0.4567
No log	20.38	265	0.6932	0.4775	0.4581
No log	20.77	270	0.6946	0.495	0.4580
No log	21.15	275	0.6931	0.5208	0.4806
No log	21.54	280	0.6931	0.5	0.4652
No log	21.92	285	0.6931	0.5067	0.4426
No log	22.31	290	0.6931	0.4908	0.3944
No log	22.69	295	0.6931	0.5317	0.4458
No log	23.08	300	0.6931	0.4933	0.4109
No log	23.46	305	0.6931	0.5025	0.4176
No log	23.85	310	0.6931	0.5033	0.4601
No log	24.23	315	0.6931	0.5017	0.4305
No log	24.62	320	0.6931	0.4883	0.4408
No log	25.0	325	0.6931	0.5267	0.4799
No log	25.38	330	0.6931	0.5083	0.4370
No log	25.77	335	0.6931	0.505	0.4396

Framework versions

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