exp2_10partition_modelo_asl6000

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: 0.1290
• Model Preparation Time: 0.0161
• Bleu Msl: 0
• Bleu 1 Msl: 0
• Bleu 2 Msl: 0
• Bleu 3 Msl: 0
• Bleu 4 Msl: 0
• Ter Msl: 100
• Bleu Asl: 0
• Bleu 1 Asl: 0.9706
• Bleu 2 Asl: 0.9514
• Bleu 3 Asl: 0.9300
• Bleu 4 Asl: 0.9044
• Ter Asl: 3.5107

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	0.1434	0.0161	0	0	0	0	0	100	0	0.9561	0.9283	0.8952	0.8607	5.5112
No log	2.0	300	0.1154	0.0161	0	0	0	0	0	100	0	0.9666	0.9439	0.9165	0.8862	4.0609
No log	3.0	450	0.0926	0.0161	0	0	0	0	0	100	0	0.9762	0.9586	0.9350	0.9090	2.9732
0.2619	4.0	600	0.0961	0.0161	0	0	0	0	0	100	0	0.9733	0.9544	0.9307	0.9045	3.2995
0.2619	5.0	750	0.1004	0.0161	0	0	0	0	0	100	0	0.9768	0.9592	0.9360	0.9096	3.0094
0.2619	6.0	900	0.0940	0.0161	0	0	0	0	0	100	0	0.9778	0.9617	0.9404	0.9165	2.7556
0.033	7.0	1050	0.0980	0.0161	0	0	0	0	0	100	0	0.9771	0.9604	0.9388	0.9147	2.7919
0.033	8.0	1200	0.1034	0.0161	0	0	0	0	0	100	0	0.9781	0.9623	0.9413	0.9178	2.8281
0.033	9.0	1350	0.0879	0.0161	0	0	0	0	0	100	0	0.9797	0.9642	0.9435	0.9199	2.6468
0.0138	10.0	1500	0.0999	0.0161	0	0	0	0	0	100	0	0.9756	0.9587	0.9373	0.9132	2.9369
0.0138	11.0	1650	0.0959	0.0161	0	0	0	0	0	100	0	0.9781	0.9612	0.9394	0.9155	2.8281
0.0138	12.0	1800	0.1083	0.0161	0	0	0	0	0	100	0	0.9774	0.9613	0.9404	0.9164	2.9007
0.0138	13.0	1950	0.1005	0.0161	0	0	0	0	0	100	0	0.9722	0.9556	0.9339	0.9089	3.5896
0.0097	14.0	2100	0.0967	0.0161	0	0	0	0	0	100	0	0.9778	0.9626	0.9419	0.9181	2.6106
0.0097	15.0	2250	0.0918	0.0161	0	0	0	0	0	100	0	0.9781	0.9621	0.9409	0.9174	2.7556
0.0097	16.0	2400	0.0904	0.0161	0	0	0	0	0	100	0	0.9784	0.9639	0.9439	0.9211	2.6831
0.0071	17.0	2550	0.0897	0.0161	0	0	0	0	0	100	0	0.9822	0.9689	0.9499	0.9284	2.1392
0.0071	18.0	2700	0.0864	0.0161	0	0	0	0	0	100	0	0.9835	0.9709	0.9524	0.9315	2.0667
0.0071	19.0	2850	0.0885	0.0161	0	0	0	0	0	100	0	0.9809	0.9666	0.9474	0.9256	2.3568
0.0041	20.0	3000	0.0905	0.0161	0	0	0	0	0	100	0	0.9828	0.9699	0.9522	0.9321	2.1755
0.0041	21.0	3150	0.0923	0.0161	0	0	0	0	0	100	0	0.9822	0.9688	0.9504	0.9299	2.2843
0.0041	22.0	3300	0.0936	0.0161	0	0	0	0	0	100	0	0.9819	0.9683	0.9498	0.9288	2.2480
0.0041	23.0	3450	0.0932	0.0161	0	0	0	0	0	100	0	0.9810	0.9673	0.9482	0.9266	2.3205
0.0034	24.0	3600	0.0923	0.0161	0	0	0	0	0	100	0	0.9816	0.9684	0.9497	0.9288	2.2480
0.0034	25.0	3750	0.0931	0.0161	0	0	0	0	0	100	0	0.9810	0.9673	0.9480	0.9260	2.3205
0.0034	26.0	3900	0.0936	0.0161	0	0	0	0	0	100	0	0.9819	0.9687	0.9499	0.9288	2.2117
0.0016	27.0	4050	0.0927	0.0161	0	0	0	0	0	100	0	0.9816	0.9686	0.9500	0.9290	2.2480
0.0016	28.0	4200	0.0927	0.0161	0	0	0	0	0	100	0	0.9816	0.9686	0.9500	0.9290	2.2480
0.0016	29.0	4350	0.0929	0.0161	0	0	0	0	0	100	0	0.9816	0.9686	0.9500	0.9290	2.2480
0.0019	30.0	4500	0.0931	0.0161	0	0	0	0	0	100	0	0.9816	0.9686	0.9500	0.9290	2.2480

Framework versions

• Transformers 4.50.2
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1