mdeberta-v3-base_ordinal_5_seed69_multilingual

This model is a fine-tuned version of microsoft/mdeberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.2865
• Mse: 2.5355
• Rmse: 1.5923
• Mae: 1.0336
• R2: 0.2365
• F1: 0.7527
• Precision: 0.7544
• Recall: 0.7591
• Accuracy: 0.7591

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-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Mse	Rmse	Mae	R2	F1	Precision	Recall	Accuracy
3.4461	0.1449	100	3.2319	3.3564	1.8321	1.7300	-0.0820	0.5372	0.4484	0.6697	0.6697
3.1709	0.2899	200	2.9852	3.1868	1.7852	1.5359	-0.0273	0.5372	0.4484	0.6697	0.6697
3.0364	0.4348	300	2.8881	3.2227	1.7952	1.5359	-0.0389	0.5372	0.4484	0.6697	0.6697
2.8969	0.5797	400	2.7693	3.0057	1.7337	1.3271	0.0311	0.5372	0.4484	0.6697	0.6697
2.8529	0.7246	500	2.6898	2.8148	1.6777	1.3434	0.0926	0.5372	0.4484	0.6697	0.6697
2.781	0.8696	600	2.6252	2.7806	1.6675	1.3010	0.1036	0.5372	0.4484	0.6697	0.6697
2.6867	1.0145	700	2.5549	2.8711	1.6944	1.2088	0.0744	0.5893	0.6724	0.6827	0.6827
2.4738	1.1594	800	2.5246	2.8687	1.6937	1.1574	0.0752	0.7006	0.6990	0.7129	0.7129
2.4723	1.3043	900	2.4213	2.6020	1.6131	1.1126	0.1612	0.6983	0.7216	0.7300	0.7300
2.5362	1.4493	1000	2.4109	2.5506	1.5971	1.1884	0.1778	0.7136	0.7133	0.7259	0.7259
2.4032	1.5942	1100	2.4424	2.7520	1.6589	1.0799	0.1128	0.7040	0.7158	0.7284	0.7284
2.3445	1.7391	1200	2.3788	2.6387	1.6244	1.0954	0.1494	0.7336	0.7315	0.7390	0.7390
2.364	1.8841	1300	2.3767	2.6533	1.6289	1.0595	0.1446	0.7150	0.7298	0.7390	0.7390
2.3271	2.0290	1400	2.4050	2.7684	1.6638	1.1060	0.1076	0.7180	0.7161	0.7268	0.7268
2.1281	2.1739	1500	2.4391	2.7920	1.6709	1.0775	0.0999	0.7238	0.7231	0.7341	0.7341
2.1314	2.3188	1600	2.3782	2.6688	1.6337	1.0799	0.1396	0.7338	0.7323	0.7357	0.7357
2.1279	2.4638	1700	2.3962	2.7325	1.6530	1.0538	0.1191	0.7227	0.7335	0.7431	0.7431
2.0826	2.6087	1800	2.3376	2.6142	1.6168	1.0449	0.1573	0.7335	0.7351	0.7455	0.7455
2.1644	2.7536	1900	2.3175	2.5522	1.5976	1.0791	0.1772	0.7312	0.7293	0.7341	0.7341
2.0112	2.8986	2000	2.2738	2.5220	1.5881	1.0261	0.1870	0.7459	0.7452	0.7537	0.7537
1.9844	3.0435	2100	2.3487	2.6860	1.6389	1.0122	0.1341	0.7405	0.7410	0.7504	0.7504
1.8639	3.1884	2200	2.3881	2.7716	1.6648	1.0082	0.1065	0.7447	0.7431	0.7471	0.7471
1.7698	3.3333	2300	2.4045	2.8018	1.6739	0.9927	0.0968	0.7484	0.7465	0.7520	0.7520

Framework versions

• Transformers 4.40.2
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.19.1