barto_exp1_10partition_modelo_msl6000

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: 0.9532
• Model Preparation Time: 0.0034
• Bleu Msl: 0
• Bleu 1 Msl: 0.7639
• Bleu 2 Msl: 0.7149
• Bleu 3 Msl: 0.6525
• Bleu 4 Msl: 0.5443
• Ter Msl: 28.3847
• Bleu Asl: 0
• Bleu 1 Asl: 0.9293
• Bleu 2 Asl: 0.8946
• Bleu 3 Asl: 0.8600
• Bleu 4 Asl: 0.8197
• Ter Asl: 9.1654

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	150	1.2803	0.0034	0	0.6793	0.5456	0.3942	0.2008	47.0772	0	0.8508	0.7752	0.7072	0.6397	18.5776
No log	2.0	300	1.0222	0.0034	0	0.7352	0.6454	0.5466	0.4151	42.4843	0	0.8848	0.8238	0.7679	0.7111	14.2235
No log	3.0	450	0.9065	0.0034	0	0.7557	0.6691	0.5770	0.4421	37.6827	0	0.9055	0.8516	0.8027	0.7522	11.9013
1.7579	4.0	600	0.9370	0.0034	0	0.7106	0.6262	0.5307	0.3905	43.8413	0	0.9287	0.8872	0.8472	0.8047	9.0711
1.7579	5.0	750	0.8798	0.0034	0	0.744	0.6483	0.5455	0.3693	37.6827	0	0.9301	0.8896	0.8513	0.8104	8.8534
1.7579	6.0	900	0.8268	0.0034	0	0.7777	0.6941	0.5919	0.4367	32.2547	0	0.9348	0.8988	0.8631	0.8230	8.5631
0.2488	7.0	1050	0.8773	0.0034	0	0.7458	0.6463	0.5394	0.3955	39.7704	0	0.9305	0.8931	0.8572	0.8183	8.7808
0.2488	8.0	1200	0.8616	0.0034	0	0.7766	0.6871	0.5825	0.4373	33.9248	0	0.9417	0.9095	0.8787	0.8464	7.2569
0.2488	9.0	1350	0.8660	0.0034	0	0.7496	0.6521	0.5392	0.3802	39.3528	0	0.9429	0.9116	0.8811	0.8464	7.1118
0.1035	10.0	1500	0.9077	0.0034	0	0.7380	0.6450	0.5310	0.3551	39.8747	0	0.9427	0.9122	0.8840	0.8534	7.4746
0.1035	11.0	1650	0.9008	0.0034	0	0.6960	0.6035	0.5006	0.3589	40.7098	0	0.9427	0.9110	0.8808	0.8483	7.5472
0.1035	12.0	1800	0.9089	0.0034	0	0.7222	0.6306	0.5310	0.3713	40.7098	0	0.9422	0.9117	0.8820	0.8496	7.6197
0.1035	13.0	1950	0.8632	0.0034	0	0.7502	0.6575	0.5511	0.4091	38.5177	0	0.9416	0.9090	0.8776	0.8423	7.6923
0.0495	14.0	2100	0.9132	0.0034	0	0.6990	0.5972	0.4719	0.3275	45.6159	0	0.9326	0.9003	0.8683	0.8325	8.5631
0.0495	15.0	2250	0.8947	0.0034	0	0.6974	0.6036	0.4808	0.3461	44.1545	0	0.9409	0.9097	0.8803	0.8484	7.9826
0.0495	16.0	2400	0.9675	0.0034	0	0.6752	0.5754	0.4583	0.3035	47.7035	0	0.9398	0.9100	0.8812	0.8490	7.8374
0.0279	17.0	2550	0.8933	0.0034	0	0.6144	0.5142	0.4073	0.2809	52.0877	0	0.9386	0.9080	0.8788	0.8462	7.6923
0.0279	18.0	2700	0.9261	0.0034	0	0.6965	0.5974	0.4745	0.3043	45.3027	0	0.9356	0.9043	0.8748	0.8428	8.1277
0.0279	19.0	2850	0.9349	0.0034	0	0.6877	0.5828	0.4633	0.3246	47.2860	0	0.9422	0.9120	0.8831	0.8524	7.5472
0.0179	20.0	3000	0.9240	0.0034	0	0.7345	0.6479	0.5332	0.3636	40.6054	0	0.9418	0.9131	0.8855	0.8546	7.2569
0.0179	21.0	3150	0.9256	0.0034	0	0.7046	0.5958	0.4713	0.3168	44.0501	0	0.9434	0.9137	0.8849	0.8531	7.2569
0.0179	22.0	3300	0.9344	0.0034	0	0.7389	0.6511	0.5386	0.3806	37.8914	0	0.9371	0.9054	0.8743	0.8406	8.0552
0.0179	23.0	3450	0.9214	0.0034	0	0.7286	0.6367	0.5203	0.3562	42.5887	0	0.9389	0.9089	0.8807	0.8493	7.7649
0.0111	24.0	3600	0.9181	0.0034	0	0.7315	0.6346	0.5186	0.3594	40.3967	0	0.9400	0.9101	0.8810	0.8494	7.7649
0.0111	25.0	3750	0.8888	0.0034	0	0.7639	0.6738	0.5603	0.3930	38.9353	0	0.9477	0.9188	0.8904	0.8584	6.6763
0.0111	26.0	3900	0.9291	0.0034	0	0.6978	0.5936	0.4724	0.3117	45.0939	0	0.9370	0.9055	0.8757	0.8434	8.0552
0.0093	27.0	4050	0.9178	0.0034	0	0.7327	0.6379	0.5191	0.3538	41.3361	0	0.9405	0.9109	0.8833	0.8531	7.5472
0.0093	28.0	4200	0.9152	0.0034	0	0.7329	0.6373	0.5154	0.3502	41.7537	0	0.9458	0.9166	0.8893	0.8591	7.1843
0.0093	29.0	4350	0.9204	0.0034	0	0.7399	0.6449	0.5254	0.3587	41.1273	0	0.9418	0.9124	0.8844	0.8535	7.4746
0.0066	30.0	4500	0.9122	0.0034	0	0.7397	0.6463	0.5293	0.3629	41.0230	0	0.9465	0.9179	0.8902	0.8600	6.8940

Framework versions

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