exp4_10partition_modelo_msl3000

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: 2.0037
• Model Preparation Time: 0.0037
• Bleu Msl: 0
• Bleu 1 Msl: 0.5874
• Bleu 2 Msl: 0.4646
• Bleu 3 Msl: 0.3297
• Bleu 4 Msl: 0.1962
• Ter Msl: 46.5481
• Bleu Asl: 0
• Bleu 1 Asl: 0
• Bleu 2 Asl: 0
• Bleu 3 Asl: 0
• Bleu 4 Asl: 0
• Ter Asl: 100

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	75	2.9399	0.0037	0	0.3001	0.1622	0.0842	0.0188	96.5412	0	0	0	0	0	100
No log	2.0	150	1.9785	0.0037	0	0.5430	0.4254	0.3242	0.2075	58.6979	0	0	0	0	0	100
No log	3.0	225	1.7198	0.0037	0	0.5523	0.4494	0.3476	0.2320	54.0183	0	0	0	0	0	100
No log	4.0	300	1.5583	0.0037	0	0.6221	0.5243	0.4176	0.2868	46.8973	0	0	0	0	0	100
No log	5.0	375	1.4945	0.0037	0	0.5987	0.4987	0.3973	0.2722	50.7630	0	0	0	0	0	100
No log	6.0	450	1.4265	0.0037	0	0.6400	0.5466	0.4332	0.2973	45.2696	0	0	0	0	0	100
1.1373	7.0	525	1.4256	0.0037	0	0.6417	0.5536	0.4496	0.3241	45.0661	0	0	0	0	0	100
1.1373	8.0	600	1.5574	0.0037	0	0.4878	0.3967	0.3028	0.1932	59.7152	0	0	0	0	0	100
1.1373	9.0	675	1.6346	0.0037	0	0.6273	0.5435	0.4374	0.3082	46.9990	0	0	0	0	0	100
1.1373	10.0	750	1.5622	0.0037	0	0.5987	0.5046	0.3941	0.2560	49.8474	0	0	0	0	0	100
1.1373	11.0	825	1.6388	0.0037	0	0.6192	0.5227	0.4073	0.2703	46.2869	0	0	0	0	0	100
1.1373	12.0	900	1.6751	0.0037	0	0.5416	0.4493	0.3522	0.2400	54.8321	0	0	0	0	0	100
1.1373	13.0	975	1.5463	0.0037	0	0.5942	0.4971	0.3952	0.2553	47.8128	0	0	0	0	0	100
0.0726	14.0	1050	1.6017	0.0037	0	0.6017	0.5151	0.4147	0.2897	47.8128	0	0	0	0	0	100
0.0726	15.0	1125	1.6789	0.0037	0	0.5322	0.4453	0.3466	0.2221	55.0356	0	0	0	0	0	100
0.0726	16.0	1200	1.6566	0.0037	0	0.5619	0.4685	0.3696	0.2460	52.1872	0	0	0	0	0	100
0.0726	17.0	1275	1.5838	0.0037	0	0.6017	0.5063	0.4066	0.2831	47.7111	0	0	0	0	0	100
0.0726	18.0	1350	1.6940	0.0037	0	0.5629	0.4681	0.3665	0.2448	50.3561	0	0	0	0	0	100
0.0726	19.0	1425	1.6916	0.0037	0	0.5933	0.5027	0.3959	0.2689	47.6094	0	0	0	0	0	100
0.0272	20.0	1500	1.7227	0.0037	0	0.6135	0.5249	0.4189	0.2844	45.8800	0	0	0	0	0	100
0.0272	21.0	1575	1.6682	0.0037	0	0.5776	0.4868	0.3862	0.2677	49.3388	0	0	0	0	0	100
0.0272	22.0	1650	1.7365	0.0037	0	0.5592	0.4636	0.3606	0.2448	52.8993	0	0	0	0	0	100
0.0272	23.0	1725	1.6694	0.0037	0	0.6024	0.5128	0.4127	0.2895	48.4232	0	0	0	0	0	100
0.0272	24.0	1800	1.7002	0.0037	0	0.6221	0.5349	0.4333	0.3062	45.5748	0	0	0	0	0	100
0.0272	25.0	1875	1.7259	0.0037	0	0.5776	0.4853	0.3841	0.2675	49.6439	0	0	0	0	0	100
0.0272	26.0	1950	1.6636	0.0037	0	0.5968	0.5053	0.4056	0.2857	47.2024	0	0	0	0	0	100
0.0147	27.0	2025	1.6579	0.0037	0	0.5858	0.4963	0.4007	0.2811	48.5249	0	0	0	0	0	100
0.0147	28.0	2100	1.6621	0.0037	0	0.6023	0.5134	0.4170	0.2958	47.2024	0	0	0	0	0	100
0.0147	29.0	2175	1.6655	0.0037	0	0.6050	0.5132	0.4144	0.2912	47.4059	0	0	0	0	0	100
0.0147	30.0	2250	1.6620	0.0037	0	0.6032	0.5113	0.4138	0.2918	47.5076	0	0	0	0	0	100

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1