mdeberta-v3-base_regression_5_seed420_NL-IT

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: 0.7614
• Mse: 2.8217
• Rmse: 1.6798
• Mae: 1.0340
• R2: 0.1242
• F1: 0.7343
• Precision: 0.7402
• Recall: 0.7466
• Accuracy: 0.7466

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
1.2168	0.2105	100	1.0410	3.8466	1.9613	1.4687	-0.2433	0.5333	0.4444	0.6667	0.6667
1.053	0.4211	200	1.0115	3.3031	1.8174	1.4738	-0.0676	0.5333	0.4444	0.6667	0.6667
1.0263	0.6316	300	0.9647	2.7400	1.6553	1.4206	0.1144	0.5333	0.4444	0.6667	0.6667
0.9754	0.8421	400	0.9284	3.0990	1.7604	1.3448	-0.0017	0.5333	0.4444	0.6667	0.6667
0.934	1.0526	500	0.8970	2.9576	1.7198	1.3110	0.0440	0.5388	0.7788	0.6690	0.6690
0.8667	1.2632	600	0.8632	2.8116	1.6768	1.2718	0.0912	0.6366	0.7004	0.7011	0.7011
0.8634	1.4737	700	0.8375	2.7817	1.6679	1.2347	0.1009	0.6477	0.6730	0.6940	0.6940
0.8294	1.6842	800	0.8966	3.0513	1.7468	1.2882	0.0137	0.6862	0.6831	0.6916	0.6916
0.819	1.8947	900	0.9153	3.1499	1.7748	1.3015	-0.0181	0.6861	0.6845	0.6880	0.6880
0.7417	2.1053	1000	0.8207	2.8460	1.6870	1.1782	0.0801	0.6577	0.6891	0.7034	0.7034
0.7342	2.3158	1100	0.8174	2.8473	1.6874	1.1696	0.0797	0.6961	0.7058	0.7200	0.7200
0.695	2.5263	1200	0.8344	2.9407	1.7149	1.1834	0.0495	0.7104	0.7086	0.7129	0.7129
0.7682	2.7368	1300	0.8055	2.8003	1.6734	1.1563	0.0949	0.7252	0.7258	0.7367	0.7367
0.702	2.9474	1400	0.7758	2.6921	1.6408	1.1185	0.1298	0.7143	0.7198	0.7319	0.7319
0.6973	3.1579	1500	0.7973	2.8367	1.6842	1.1395	0.0831	0.7346	0.7332	0.7367	0.7367
0.6206	3.3684	1600	0.7865	2.7933	1.6713	1.1160	0.0971	0.7216	0.7240	0.7355	0.7355
0.6859	3.5789	1700	0.7750	2.7686	1.6639	1.1000	0.1051	0.7081	0.7257	0.7343	0.7343
0.6493	3.7895	1800	0.7721	2.7292	1.6520	1.0992	0.1178	0.7145	0.7313	0.7390	0.7390
0.6285	4.0	1900	0.8107	2.8467	1.6872	1.1415	0.0799	0.7194	0.7186	0.7295	0.7295
0.5887	4.2105	2000	0.8451	3.0451	1.7450	1.1710	0.0158	0.7240	0.7309	0.7200	0.7200
0.6098	4.4211	2100	0.7592	2.6481	1.6273	1.0817	0.1441	0.7194	0.7289	0.7390	0.7390
0.5907	4.6316	2200	0.7595	2.7178	1.6486	1.0643	0.1215	0.7230	0.7334	0.7426	0.7426
0.5555	4.8421	2300	0.7761	2.7759	1.6661	1.0820	0.1028	0.7304	0.7289	0.7378	0.7378
0.6021	5.0526	2400	0.7987	2.8809	1.6973	1.1033	0.0688	0.7221	0.7202	0.7295	0.7295
0.5504	5.2632	2500	0.7843	2.8168	1.6783	1.0895	0.0895	0.7370	0.7352	0.7426	0.7426
0.5052	5.4737	2600	0.7873	2.8846	1.6984	1.0834	0.0676	0.7401	0.7417	0.7509	0.7509
0.5171	5.6842	2700	0.7808	2.8328	1.6831	1.0866	0.0844	0.7317	0.7297	0.7367	0.7367
0.5395	5.8947	2800	0.7652	2.7540	1.6595	1.0682	0.1098	0.7322	0.7305	0.7390	0.7390
0.5247	6.1053	2900	0.7771	2.8384	1.6848	1.0703	0.0826	0.7256	0.7281	0.7390	0.7390
0.4707	6.3158	3000	0.8009	2.9554	1.7191	1.0902	0.0447	0.7231	0.7208	0.7284	0.7284
0.5139	6.5263	3100	0.7848	2.9021	1.7035	1.0748	0.0620	0.7360	0.7357	0.7450	0.7450
0.4924	6.7368	3200	0.7731	2.8285	1.6818	1.0634	0.0857	0.7182	0.7337	0.7414	0.7414
0.4907	6.9474	3300	0.7731	2.8268	1.6813	1.0574	0.0863	0.7209	0.7268	0.7378	0.7378
0.4836	7.1579	3400	0.7811	2.8490	1.6879	1.0718	0.0791	0.7252	0.7236	0.7331	0.7331
0.458	7.3684	3500	0.7863	2.9145	1.7072	1.0651	0.0580	0.7186	0.7201	0.7319	0.7319
0.4281	7.5789	3600	0.7782	2.8838	1.6982	1.0606	0.0679	0.7388	0.7377	0.7461	0.7461
0.4267	7.7895	3700	0.7914	2.9346	1.7131	1.0837	0.0515	0.7438	0.7452	0.7426	0.7426
0.474	8.0	3800	0.7600	2.7846	1.6687	1.0396	0.1000	0.7337	0.7350	0.7450	0.7450
0.4033	8.2105	3900	0.7654	2.8418	1.6858	1.0383	0.0815	0.7270	0.7357	0.7450	0.7450
0.4517	8.4211	4000	0.7807	2.9020	1.7035	1.0540	0.0620	0.7193	0.7239	0.7355	0.7355
0.4657	8.6316	4100	0.7809	2.8977	1.7023	1.0572	0.0634	0.7178	0.7212	0.7331	0.7331
0.4225	8.8421	4200	0.7971	2.9923	1.7298	1.0833	0.0328	0.7354	0.7338	0.7378	0.7378
0.4221	9.0526	4300	0.7862	2.9421	1.7153	1.0688	0.0490	0.7256	0.7238	0.7331	0.7331

Framework versions

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