exp1_10partition_modelo9000

This model is a fine-tuned version of Helsinki-NLP/opus-mt-es-es on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7492
• Model Preparation Time: 0.0049
• Bleu Msl: 0
• Bleu 1 Msl: 0.7877
• Bleu 2 Msl: 0.6701
• Bleu 3 Msl: 0.5934
• Bleu 4 Msl: 0.5057
• Ter Msl: 31.9909
• Bleu Asl: 0
• Bleu 1 Asl: 0.9668
• Bleu 2 Asl: 0.9490
• Bleu 3 Asl: 0.9272
• Bleu 4 Asl: 0.8987
• Ter Asl: 4.1176

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	225	0.3414	0.0049	0	0.8217	0.7593	0.6876	0.5664	23.2578	0	0.9399	0.9041	0.8675	0.8270	7.4288
No log	2.0	450	0.2770	0.0049	0	0.8527	0.7879	0.7104	0.5930	21.7464	0	0.9581	0.9312	0.9040	0.8735	5.0126
0.4738	3.0	675	0.2653	0.0049	0	0.8642	0.8091	0.7464	0.6360	19.8153	0	0.9675	0.9457	0.9214	0.8928	4.0029
0.4738	4.0	900	0.2708	0.0049	0	0.8372	0.7858	0.7228	0.6126	21.2427	0	0.9604	0.9360	0.9109	0.8823	4.7241
0.0823	5.0	1125	0.2615	0.0049	0	0.8642	0.8075	0.7488	0.6515	20.3191	0	0.9645	0.9431	0.9203	0.8936	4.1832
0.0823	6.0	1350	0.2768	0.0049	0	0.7286	0.6666	0.5956	0.4899	41.3098	0	0.8344	0.8029	0.7687	0.7246	22.2863
0.0441	7.0	1575	0.2917	0.0049	0	0.8412	0.7835	0.7112	0.5945	20.9068	0	0.9678	0.9467	0.9242	0.8976	4.1111
0.0441	8.0	1800	0.2729	0.0049	0	0.8458	0.7909	0.7301	0.6287	18.7238	0	0.9704	0.9508	0.9295	0.9041	3.5701
0.0294	9.0	2025	0.2944	0.0049	0	0.8404	0.7712	0.6997	0.5950	21.9144	0	0.9685	0.9488	0.9277	0.9028	3.7505
0.0294	10.0	2250	0.2702	0.0049	0	0.8552	0.7927	0.7219	0.6148	19.6474	0	0.9655	0.9457	0.9247	0.9001	4.0750
0.0294	11.0	2475	0.2753	0.0049	0	0.8563	0.7929	0.7218	0.6159	21.2427	0	0.9688	0.9485	0.9260	0.8998	3.7865
0.0214	12.0	2700	0.2631	0.0049	0	0.8507	0.7936	0.7298	0.6288	19.1436	0	0.9704	0.9507	0.9290	0.9042	3.6423
0.0214	13.0	2925	0.2897	0.0049	0	0.8526	0.7939	0.7308	0.6291	19.3955	0	0.9655	0.9462	0.9246	0.8988	4.0029
0.0142	14.0	3150	0.2763	0.0049	0	0.8414	0.7851	0.7219	0.6247	20.5709	0	0.9686	0.9490	0.9273	0.9024	3.7144
0.0142	15.0	3375	0.2882	0.0049	0	0.8461	0.7808	0.7112	0.6074	20.2351	0	0.9745	0.9572	0.9378	0.9149	3.1735
0.0113	16.0	3600	0.2712	0.0049	0	0.8526	0.7883	0.7111	0.5978	19.5634	0	0.9738	0.9567	0.9368	0.9134	3.2456
0.0113	17.0	3825	0.2899	0.0049	0	0.8459	0.7788	0.7053	0.5981	19.8153	0	0.9723	0.9540	0.9343	0.9111	3.3898
0.0094	18.0	4050	0.2962	0.0049	0	0.8429	0.7788	0.7044	0.5902	21.0747	0	0.9701	0.9519	0.9315	0.9077	3.6423
0.0094	19.0	4275	0.2855	0.0049	0	0.8596	0.7984	0.7262	0.6185	19.2275	0	0.9754	0.9580	0.9392	0.9168	2.9931
0.0075	20.0	4500	0.2864	0.0049	0	0.8507	0.7936	0.7277	0.6206	19.8153	0	0.9699	0.9518	0.9324	0.9092	3.5701
0.0075	21.0	4725	0.2843	0.0049	0	0.8509	0.7880	0.7164	0.6099	20.6549	0	0.9742	0.9573	0.9388	0.9165	3.2095
0.0075	22.0	4950	0.3010	0.0049	0	0.8507	0.7911	0.7223	0.6177	21.3266	0	0.9738	0.9561	0.9369	0.9140	3.1735
0.0058	23.0	5175	0.2967	0.0049	0	0.8478	0.7907	0.7218	0.6155	20.7389	0	0.9723	0.9542	0.9349	0.9117	3.2816
0.0058	24.0	5400	0.2965	0.0049	0	0.8516	0.7908	0.7202	0.6111	21.2427	0	0.9754	0.9580	0.9392	0.9165	2.9931
0.0052	25.0	5625	0.2902	0.0049	0	0.8553	0.7922	0.7211	0.6146	21.1587	0	0.8396	0.8118	0.7820	0.7425	21.3487
0.0052	26.0	5850	0.2839	0.0049	0	0.8546	0.7997	0.7348	0.6313	19.5634	0	0.9723	0.9534	0.9337	0.9106	3.3177
0.0052	27.0	6075	0.2886	0.0049	0	0.8460	0.7860	0.7179	0.6125	20.6549	0	0.9745	0.9570	0.9380	0.9152	3.1013
0.0052	28.0	6300	0.2930	0.0049	0	0.8519	0.7882	0.7171	0.6100	20.9068	0	0.9732	0.9557	0.9363	0.9134	3.2816
0.0038	29.0	6525	0.2905	0.0049	0	0.8517	0.7905	0.7208	0.6141	20.8228	0	0.9732	0.9557	0.9363	0.9133	3.2456
0.0038	30.0	6750	0.2905	0.0049	0	0.8517	0.7905	0.7208	0.6141	20.8228	0	0.9735	0.9562	0.9372	0.9145	3.2095

Framework versions

• Transformers 4.50.0
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.21.1