exp4_10partition_modelo3000

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: 1.5196
• Model Preparation Time: 0.0032
• Bleu Msl: 0
• Bleu 1 Msl: 0.7743
• Bleu 2 Msl: 0.6636
• Bleu 3 Msl: 0.5100
• Bleu 4 Msl: 0.3308
• Ter Msl: 29.7071
• 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	1.7039	0.0032	0	0.1241	0.0887	0.0609	0.0353	186.5717	0	0	0	0	0	100
No log	2.0	150	1.2533	0.0032	0	0.0633	0.0461	0.0327	0.0198	374.9746	0	0	0	0	0	100
No log	3.0	225	1.1708	0.0032	0	0.7083	0.5967	0.4821	0.3334	35.0966	0	0	0	0	0	100
No log	4.0	300	1.1708	0.0032	0	0.6275	0.5313	0.4212	0.2860	33.6724	0	0	0	0	0	100
No log	5.0	375	1.2124	0.0032	0	0.6280	0.5237	0.4195	0.2871	35.4018	0	0	0	0	0	100
No log	6.0	450	1.2659	0.0032	0	0.7330	0.6390	0.5212	0.3616	32.2482	0	0	0	0	0	100
0.5393	7.0	525	1.2558	0.0032	0	0.7346	0.6547	0.5445	0.3905	31.2309	0	0	0	0	0	100
0.5393	8.0	600	1.2707	0.0032	0	0.7318	0.6494	0.5454	0.3867	31.7396	0	0	0	0	0	100
0.5393	9.0	675	1.2536	0.0032	0	0.7497	0.6642	0.5502	0.3825	30.0102	0	0	0	0	0	100
0.5393	10.0	750	1.1947	0.0032	0	0.7477	0.6674	0.5470	0.3748	29.8067	0	0	0	0	0	100
0.5393	11.0	825	1.3127	0.0032	0	0.7238	0.6425	0.5318	0.3704	31.6378	0	0	0	0	0	100
0.5393	12.0	900	1.2285	0.0032	0	0.7659	0.6718	0.5562	0.3969	29.0946	0	0	0	0	0	100
0.5393	13.0	975	1.2194	0.0032	0	0.7558	0.6752	0.5585	0.4031	28.6877	0	0	0	0	0	100
0.0401	14.0	1050	1.2710	0.0032	0	0.7419	0.6661	0.5555	0.3996	30.2136	0	0	0	0	0	100
0.0401	15.0	1125	1.2230	0.0032	0	0.7714	0.6850	0.5688	0.4073	27.5687	0	0	0	0	0	100
0.0401	16.0	1200	1.2397	0.0032	0	0.7654	0.6854	0.5685	0.4114	28.9929	0	0	0	0	0	100
0.0401	17.0	1275	1.2646	0.0032	0	0.7694	0.6900	0.5772	0.4145	27.6704	0	0	0	0	0	100
0.0401	18.0	1350	1.2835	0.0032	0	0.7618	0.6865	0.5728	0.4085	28.1790	0	0	0	0	0	100
0.0401	19.0	1425	1.2831	0.0032	0	0.7568	0.6809	0.5669	0.4015	28.6877	0	0	0	0	0	100
0.0189	20.0	1500	1.2742	0.0032	0	0.7632	0.6812	0.5609	0.3968	28.8911	0	0	0	0	0	100
0.0189	21.0	1575	1.2997	0.0032	0	0.7572	0.6774	0.5635	0.3996	29.8067	0	0	0	0	0	100
0.0189	22.0	1650	1.2868	0.0032	0	0.7641	0.6860	0.5734	0.4094	28.4842	0	0	0	0	0	100
0.0189	23.0	1725	1.2870	0.0032	0	0.7630	0.6850	0.5718	0.4083	28.4842	0	0	0	0	0	100
0.0189	24.0	1800	1.3061	0.0032	0	0.7626	0.6904	0.5810	0.4212	28.1790	0	0	0	0	0	100
0.0189	25.0	1875	1.3040	0.0032	0	0.7621	0.6843	0.5724	0.4111	29.1963	0	0	0	0	0	100
0.0189	26.0	1950	1.2923	0.0032	0	0.7641	0.6892	0.5782	0.4138	28.3825	0	0	0	0	0	100
0.0129	27.0	2025	1.3121	0.0032	0	0.7635	0.6873	0.5744	0.4099	28.5860	0	0	0	0	0	100
0.0129	28.0	2100	1.3181	0.0032	0	0.7600	0.6857	0.5731	0.4072	28.4842	0	0	0	0	0	100
0.0129	29.0	2175	1.3134	0.0032	0	0.7621	0.6862	0.5728	0.4074	28.4842	0	0	0	0	0	100
0.0129	30.0	2250	1.3114	0.0032	0	0.7611	0.6850	0.5721	0.4070	28.4842	0	0	0	0	0	100

Framework versions

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