barto_exp1_10partition_modelo_msl12000

This model is a fine-tuned version of vgaraujov/bart-base-spanish on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3274
• Model Preparation Time: 0.0031
• Bleu Msl: 0
• Bleu 1 Msl: 0.7983
• Bleu 2 Msl: 0.7311
• Bleu 3 Msl: 0.6693
• Bleu 4 Msl: 0.5665
• Ter Msl: 25.3289
• Bleu Asl: 0
• Bleu 1 Asl: 0.9864
• Bleu 2 Asl: 0.9742
• Bleu 3 Asl: 0.9595
• Bleu 4 Asl: 0.9414
• Ter Asl: 1.7398

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Bleu Msl	Bleu 1 Msl	Bleu 2 Msl	Bleu 3 Msl	Bleu 4 Msl	Ter Msl	Bleu Asl	Bleu 1 Asl	Bleu 2 Asl	Bleu 3 Asl	Bleu 4 Asl	Ter Asl
No log	1.0	300	0.2151	0.0031	0	0.7679	0.6803	0.6089	0.5167	24.0921	0	0.9615	0.9355	0.9075	0.8756	4.8123
0.4301	2.0	600	0.1845	0.0031	0	0.8353	0.7778	0.7160	0.6210	18.5120	0	0.9768	0.9595	0.9390	0.9149	2.8975
0.4301	3.0	900	0.1652	0.0031	0	0.7743	0.7037	0.6406	0.5480	18.1577	0	0.9753	0.9574	0.9363	0.9106	3.0990
0.0922	4.0	1200	0.1778	0.0031	0	0.8149	0.7637	0.7139	0.6257	16.2976	0	0.9768	0.9595	0.9393	0.9151	2.8975
0.0573	5.0	1500	0.1854	0.0031	0	0.8296	0.7657	0.7028	0.6154	17.6262	0	0.9798	0.9641	0.9452	0.9220	2.4943
0.0573	6.0	1800	0.2056	0.0031	0	0.8118	0.7316	0.6512	0.5375	20.6377	0	0.9798	0.9642	0.9454	0.9226	2.3684
0.0376	7.0	2100	0.1904	0.0031	0	0.8111	0.7447	0.6831	0.5942	18.9548	0	0.9819	0.9667	0.9473	0.9235	2.2676
0.0376	8.0	2400	0.1995	0.0031	0	0.7990	0.7323	0.6681	0.5774	18.2462	0	0.9837	0.9699	0.9526	0.9311	1.9400
0.0313	9.0	2700	0.1854	0.0031	0	0.7930	0.7317	0.6723	0.5888	20.5492	0	0.9830	0.9687	0.9509	0.9286	1.9652
0.0211	10.0	3000	0.1965	0.0031	0	0.7923	0.7288	0.6682	0.5762	18.6005	0	0.9832	0.9694	0.9519	0.9303	1.9652
0.0211	11.0	3300	0.1845	0.0031	0	0.7928	0.7263	0.6630	0.5748	18.8663	0	0.9835	0.9694	0.9514	0.9292	1.9652
0.0185	12.0	3600	0.2036	0.0031	0	0.8338	0.7701	0.7111	0.6219	17.1833	0	0.9822	0.9686	0.9513	0.9295	2.0408
0.0185	13.0	3900	0.2222	0.0031	0	0.8072	0.7467	0.6880	0.5971	18.6891	0	0.9852	0.9719	0.9549	0.9338	1.8141
0.0134	14.0	4200	0.2190	0.0031	0	0.7826	0.7010	0.6305	0.5388	22.4978	0	0.9847	0.9716	0.9548	0.9339	1.8644
0.0097	15.0	4500	0.2151	0.0031	0	0.8007	0.7261	0.6617	0.5711	20.0177	0	0.9843	0.9707	0.9536	0.9321	1.8644
0.0097	16.0	4800	0.2083	0.0031	0	0.8034	0.7422	0.6843	0.6028	17.4491	0	0.9845	0.9715	0.9542	0.9324	1.8896
0.0074	17.0	5100	0.2247	0.0031	0	0.7831	0.7168	0.6572	0.5693	20.2834	0	0.9841	0.9708	0.9538	0.9326	1.9904
0.0074	18.0	5400	0.2348	0.0031	0	0.8080	0.7415	0.6793	0.5904	19.3091	0	0.9839	0.9702	0.9529	0.9310	1.9652
0.007	19.0	5700	0.2282	0.0031	0	0.8198	0.7650	0.7129	0.6271	19.3091	0	0.9841	0.9708	0.9535	0.9318	1.9400
0.0054	20.0	6000	0.2363	0.0031	0	0.7859	0.7273	0.6733	0.5908	20.8149	0	0.9854	0.9724	0.9554	0.9341	1.8141
0.0054	21.0	6300	0.2313	0.0031	0	0.8296	0.7676	0.7082	0.6170	18.2462	0	0.9860	0.9737	0.9574	0.9368	1.6881
0.0044	22.0	6600	0.2352	0.0031	0	0.7974	0.7346	0.6805	0.5993	19.3977	0	0.9843	0.9712	0.9542	0.9329	1.8141
0.0044	23.0	6900	0.2402	0.0031	0	0.7982	0.7360	0.6793	0.5948	19.3091	0	0.9845	0.9712	0.9540	0.9323	1.8393
0.0041	24.0	7200	0.2345	0.0031	0	0.8148	0.7558	0.7029	0.6164	18.6005	0	0.9852	0.9725	0.9557	0.9346	1.7889
0.0029	25.0	7500	0.2363	0.0031	0	0.8262	0.7659	0.7112	0.6293	17.2719	0	0.9843	0.9711	0.9541	0.9329	1.8896
0.0029	26.0	7800	0.2451	0.0031	0	0.8177	0.7535	0.6980	0.6197	17.4491	0	0.9841	0.9710	0.9540	0.9326	1.8644
0.0028	27.0	8100	0.2483	0.0031	0	0.8246	0.7628	0.7090	0.6236	17.8034	0	0.9854	0.9727	0.9563	0.9356	1.7637
0.0028	28.0	8400	0.2459	0.0031	0	0.8212	0.7609	0.7084	0.6276	17.8034	0	0.9845	0.9715	0.9546	0.9334	1.8393
0.0021	29.0	8700	0.2491	0.0031	0	0.8239	0.7626	0.7103	0.6280	17.8034	0	0.9850	0.9720	0.9551	0.9340	1.8141
0.002	30.0	9000	0.2524	0.0031	0	0.8270	0.7666	0.7145	0.6331	17.6262	0	0.9850	0.9720	0.9551	0.9340	1.8141

Framework versions

• Transformers 4.51.1
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1