exp1_10partition_modelo6000

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: 1.0010
• Model Preparation Time: 0.0033
• Bleu Msl: 0
• Bleu 1 Msl: 0.8065
• Bleu 2 Msl: 0.6721
• Bleu 3 Msl: 0.5806
• Bleu 4 Msl: 0.4790
• Ter Msl: 32.5988
• Bleu Asl: 0
• Bleu 1 Asl: 0.9641
• Bleu 2 Asl: 0.9412
• Bleu 3 Asl: 0.9145
• Bleu 4 Asl: 0.8831
• Ter Asl: 4.2285

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	0.5835	0.0033	0	0.6935	0.6035	0.5188	0.4103	34.7607	0	0.9353	0.8936	0.8530	0.8082	8.1229
No log	2.0	300	0.4129	0.0033	0	0.3646	0.3106	0.2595	0.1977	55.8354	0	0.9507	0.9186	0.8861	0.8483	6.2116
No log	3.0	450	0.4315	0.0033	0	0.8577	0.7996	0.7370	0.6337	21.1587	0	0.9585	0.9295	0.8997	0.8651	4.8464
0.5153	4.0	600	0.4610	0.0033	0	0.8386	0.7613	0.6808	0.5652	22.9219	0	0.9670	0.9431	0.9165	0.8864	4.2321
0.5153	5.0	750	0.4584	0.0033	0	0.8280	0.7483	0.6630	0.5411	25.2729	0	0.9597	0.9333	0.9054	0.8723	5.0512
0.5153	6.0	900	0.4727	0.0033	0	0.8435	0.7837	0.7186	0.6131	21.6625	0	0.9657	0.9441	0.9203	0.8907	4.2321
0.0707	7.0	1050	0.4436	0.0033	0	0.8654	0.7865	0.7083	0.5924	23.2578	0	0.9651	0.9435	0.9204	0.8925	4.3003
0.0707	8.0	1200	0.4461	0.0033	0	0.8324	0.7633	0.6932	0.5830	22.5861	0	0.9663	0.9448	0.9223	0.8952	4.3686
0.0707	9.0	1350	0.4476	0.0033	0	0.8587	0.7927	0.7190	0.6000	20.3191	0	0.9693	0.9499	0.9296	0.9049	3.5495
0.0358	10.0	1500	0.4232	0.0033	0	0.8534	0.7827	0.7069	0.5959	20.8228	0	0.9664	0.9443	0.9202	0.8907	4.2321
0.0358	11.0	1650	0.4483	0.0033	0	0.8746	0.8137	0.7417	0.6351	18.7238	0	0.9621	0.9387	0.9141	0.8850	4.5734
0.0358	12.0	1800	0.4454	0.0033	0	0.8793	0.8180	0.7470	0.6410	17.2124	0	0.9627	0.9388	0.9139	0.8853	4.6416
0.0358	13.0	1950	0.4644	0.0033	0	0.8699	0.8175	0.7570	0.6560	16.5407	0	0.9663	0.9451	0.9225	0.8957	4.2321
0.0212	14.0	2100	0.4508	0.0033	0	0.8626	0.7989	0.7259	0.6211	20.2351	0	0.9694	0.9488	0.9263	0.8996	3.7543
0.0212	15.0	2250	0.4740	0.0033	0	0.8725	0.8094	0.7450	0.6441	18.3879	0	0.9669	0.9450	0.9212	0.8934	4.2321
0.0212	16.0	2400	0.4574	0.0033	0	0.8785	0.8100	0.7330	0.6217	19.8153	0	0.9664	0.9439	0.9203	0.8921	4.3686
0.0157	17.0	2550	0.4653	0.0033	0	0.8699	0.8000	0.7291	0.6206	19.3115	0	0.9682	0.9470	0.9242	0.8963	4.0956
0.0157	18.0	2700	0.4822	0.0033	0	0.8717	0.8049	0.7362	0.6298	18.4719	0	0.9663	0.9448	0.9211	0.8925	4.1638
0.0157	19.0	2850	0.4778	0.0033	0	0.8730	0.8035	0.7316	0.6202	19.3955	0	0.9675	0.9453	0.9216	0.8932	4.3003
0.0115	20.0	3000	0.4797	0.0033	0	0.8718	0.8116	0.7486	0.6488	18.4719	0	0.9681	0.9469	0.9247	0.8982	4.0956
0.0115	21.0	3150	0.4745	0.0033	0	0.8664	0.8028	0.7346	0.6291	19.3955	0	0.9682	0.9463	0.9228	0.8953	4.0273
0.0115	22.0	3300	0.4819	0.0033	0	0.8708	0.8090	0.7417	0.6383	18.1360	0	0.9717	0.9517	0.9303	0.9050	3.5495
0.0115	23.0	3450	0.4813	0.0033	0	0.8754	0.8162	0.7506	0.6483	18.4719	0	0.9705	0.9505	0.9294	0.9043	3.7543
0.0089	24.0	3600	0.4740	0.0033	0	0.8723	0.8118	0.7447	0.6409	19.0596	0	0.9705	0.9507	0.9307	0.9059	3.7543
0.0089	25.0	3750	0.4721	0.0033	0	0.8801	0.8241	0.7611	0.6583	17.2964	0	0.9699	0.9493	0.9275	0.9016	3.8225
0.0089	26.0	3900	0.4704	0.0033	0	0.8722	0.8102	0.7420	0.6397	19.1436	0	0.9693	0.9483	0.9260	0.9000	3.7543
0.0065	27.0	4050	0.4775	0.0033	0	0.8722	0.8141	0.7476	0.6441	18.3879	0	0.9705	0.9504	0.9291	0.9035	3.6177
0.0065	28.0	4200	0.4794	0.0033	0	0.8730	0.8160	0.7499	0.6471	18.4719	0	0.9705	0.9504	0.9290	0.9032	3.6177
0.0065	29.0	4350	0.4798	0.0033	0	0.8737	0.8163	0.7498	0.6461	18.3039	0	0.9705	0.9500	0.9287	0.9033	3.6177
0.0058	30.0	4500	0.4809	0.0033	0	0.8721	0.8157	0.7503	0.6473	18.4719	0	0.9705	0.9504	0.9293	0.9037	3.6177

Framework versions

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