exp5_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: 1.9665
• Model Preparation Time: 0.0033
• Bleu Msl: 0
• Bleu 1 Msl: 0.5931
• Bleu 2 Msl: 0.5130
• Bleu 3 Msl: 0.4339
• Bleu 4 Msl: 0.3266
• Ter Msl: 47.5877
• 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	2.7914	0.0033	0	0.1654	0.0940	0.0494	0.0128	87.5996	0	0	0	0	0	100
No log	2.0	150	1.8028	0.0033	0	0.4191	0.2877	0.2019	0.1101	67.6705	0	0	0	0	0	100
No log	3.0	225	1.4135	0.0033	0	0.5865	0.4578	0.3409	0.2456	52.1701	0	0	0	0	0	100
No log	4.0	300	1.5030	0.0033	0	0.5721	0.4443	0.3334	0.2120	50.1329	0	0	0	0	0	100
No log	5.0	375	1.4242	0.0033	0	0.5794	0.4785	0.3862	0.2714	43.9327	0	0	0	0	0	100
No log	6.0	450	1.3609	0.0033	0	0.5858	0.4832	0.3977	0.3091	44.9070	0	0	0	0	0	100
1.2827	7.0	525	1.2037	0.0033	0	0.6417	0.5654	0.4990	0.4047	36.9353	0	0	0	0	0	100
1.2827	8.0	600	1.2053	0.0033	0	0.6609	0.5676	0.4817	0.3729	36.8468	0	0	0	0	0	100
1.2827	9.0	675	1.2424	0.0033	0	0.5820	0.4509	0.3385	0.2104	48.1842	0	0	0	0	0	100
1.2827	10.0	750	1.2929	0.0033	0	0.5659	0.4539	0.3538	0.2321	48.0957	0	0	0	0	0	100
1.2827	11.0	825	1.1589	0.0033	0	0.6350	0.5270	0.4285	0.3150	42.7812	0	0	0	0	0	100
1.2827	12.0	900	1.2968	0.0033	0	0.5602	0.4495	0.3553	0.2490	48.1842	0	0	0	0	0	100
1.2827	13.0	975	1.2490	0.0033	0	0.5767	0.4619	0.3653	0.2593	47.2099	0	0	0	0	0	100
0.0822	14.0	1050	1.4277	0.0033	0	0.6078	0.5244	0.4553	0.3644	42.3384	0	0	0	0	0	100
0.0822	15.0	1125	1.4144	0.0033	0	0.5879	0.4699	0.3753	0.2648	47.6528	0	0	0	0	0	100
0.0822	16.0	1200	1.3332	0.0033	0	0.5653	0.4617	0.3742	0.2684	49.2471	0	0	0	0	0	100
0.0822	17.0	1275	1.5239	0.0033	0	0.5890	0.4859	0.3901	0.2680	44.5527	0	0	0	0	0	100
0.0822	18.0	1350	1.3280	0.0033	0	0.6151	0.5090	0.4173	0.3072	44.0213	0	0	0	0	0	100
0.0822	19.0	1425	1.3868	0.0033	0	0.5928	0.4979	0.4103	0.2937	44.7298	0	0	0	0	0	100
0.0271	20.0	1500	1.3441	0.0033	0	0.6270	0.5304	0.4450	0.3406	41.4526	0	0	0	0	0	100
0.0271	21.0	1575	1.3344	0.0033	0	0.6260	0.5239	0.4272	0.3237	39.6811	0	0	0	0	0	100
0.0271	22.0	1650	1.3984	0.0033	0	0.6113	0.5151	0.4263	0.3290	41.0983	0	0	0	0	0	100
0.0271	23.0	1725	1.3501	0.0033	0	0.6353	0.5335	0.4476	0.3477	42.1612	0	0	0	0	0	100
0.0271	24.0	1800	1.3269	0.0033	0	0.6330	0.5341	0.4477	0.3481	40.9212	0	0	0	0	0	100
0.0271	25.0	1875	1.3610	0.0033	0	0.6220	0.5271	0.4456	0.3457	41.6298	0	0	0	0	0	100
0.0271	26.0	1950	1.3784	0.0033	0	0.6096	0.5116	0.4175	0.2975	43.0469	0	0	0	0	0	100
0.0124	27.0	2025	1.3464	0.0033	0	0.6247	0.5228	0.4319	0.3250	41.9841	0	0	0	0	0	100
0.0124	28.0	2100	1.3658	0.0033	0	0.6151	0.5130	0.4204	0.3111	42.1612	0	0	0	0	0	100
0.0124	29.0	2175	1.3549	0.0033	0	0.6274	0.5252	0.4337	0.3285	41.7183	0	0	0	0	0	100
0.0124	30.0	2250	1.3565	0.0033	0	0.6263	0.5241	0.4336	0.3290	41.3640	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