barto_exp4_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: 0.7794
• Model Preparation Time: 0.0035
• Bleu Msl: 0
• Bleu 1 Msl: 0.8321
• Bleu 2 Msl: 0.7380
• Bleu 3 Msl: 0.5998
• Bleu 4 Msl: 0.4237
• Ter Msl: 23.2218
• Bleu Asl: 0
• Bleu 1 Asl: 0.9740
• Bleu 2 Asl: 0.9554
• Bleu 3 Asl: 0.9344
• Bleu 4 Asl: 0.9093
• Ter Asl: 2.9784

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	0.4939	0.0035	0	0.7940	0.7431	0.6246	0.4697	22.9908	0	0.9380	0.9009	0.8632	0.8244	7.4242
No log	2.0	300	0.4169	0.0035	0	0.8179	0.7459	0.6315	0.4817	24.2116	0	0.9568	0.9283	0.8975	0.8634	5.2273
No log	3.0	450	0.3977	0.0035	0	0.8199	0.7461	0.6370	0.4995	22.8891	0	0.9582	0.9306	0.9002	0.8657	5.0758
0.4335	4.0	600	0.4259	0.0035	0	0.8071	0.7533	0.6364	0.4776	22.9908	0	0.9627	0.9365	0.9081	0.8754	4.4697
0.4335	5.0	750	0.3683	0.0035	0	0.8599	0.8031	0.6959	0.5524	18.2096	0	0.9665	0.9431	0.9171	0.8866	3.8636
0.4335	6.0	900	0.3889	0.0035	0	0.8316	0.7665	0.6575	0.5022	21.1597	0	0.9640	0.9401	0.9136	0.8832	4.3182
0.0653	7.0	1050	0.4314	0.0035	0	0.8489	0.7888	0.6864	0.5501	20.3459	0	0.9666	0.9437	0.9177	0.8879	3.7879
0.0653	8.0	1200	0.4212	0.0035	0	0.8579	0.7999	0.6957	0.5523	18.7182	0	0.9653	0.9414	0.9154	0.8855	4.2424
0.0653	9.0	1350	0.4298	0.0035	0	0.8530	0.7917	0.6886	0.5460	19.4303	0	0.9628	0.9387	0.9111	0.8800	4.0909
0.0374	10.0	1500	0.4524	0.0035	0	0.8184	0.7516	0.6436	0.5061	23.8047	0	0.9684	0.9457	0.9206	0.8919	3.7121
0.0374	11.0	1650	0.4429	0.0035	0	0.8392	0.7797	0.6771	0.5381	22.2787	0	0.9640	0.9395	0.9129	0.8818	4.1667
0.0374	12.0	1800	0.4376	0.0035	0	0.8479	0.7939	0.6902	0.5468	19.4303	0	0.9640	0.9403	0.9137	0.8828	4.2424
0.0374	13.0	1950	0.4262	0.0035	0	0.8447	0.7834	0.6851	0.5424	20.7528	0	0.9653	0.9426	0.9169	0.8883	4.0152
0.0233	14.0	2100	0.4827	0.0035	0	0.8478	0.7877	0.6793	0.5323	19.9390	0	0.9666	0.9445	0.9186	0.8881	3.7879
0.0233	15.0	2250	0.4489	0.0035	0	0.8459	0.7872	0.6792	0.5276	19.9390	0	0.9653	0.9434	0.9184	0.8899	3.9394
0.0233	16.0	2400	0.4538	0.0035	0	0.8579	0.7989	0.6915	0.5404	18.7182	0	0.9652	0.9421	0.9154	0.8843	3.9394
0.0152	17.0	2550	0.5196	0.0035	0	0.8339	0.7799	0.6687	0.5179	19.6338	0	0.9703	0.9483	0.9234	0.8939	3.3333
0.0152	18.0	2700	0.4844	0.0035	0	0.8416	0.7862	0.6851	0.5429	19.6338	0	0.9665	0.9440	0.9192	0.8897	3.8636
0.0152	19.0	2850	0.4768	0.0035	0	0.8498	0.7884	0.6839	0.5380	20.2442	0	0.9665	0.9436	0.9171	0.8856	3.9394
0.0111	20.0	3000	0.4795	0.0035	0	0.8599	0.8037	0.7016	0.5574	18.5148	0	0.9652	0.9413	0.9142	0.8823	4.0909
0.0111	21.0	3150	0.4967	0.0035	0	0.8400	0.7771	0.6712	0.5224	20.3459	0	0.9690	0.9467	0.9211	0.8912	3.7121
0.0111	22.0	3300	0.4832	0.0035	0	0.8518	0.7905	0.6817	0.5365	18.9217	0	0.9678	0.9449	0.9189	0.8882	3.7879
0.0111	23.0	3450	0.5077	0.0035	0	0.8479	0.7921	0.6842	0.5358	18.9217	0	0.9665	0.9431	0.9169	0.8858	3.8636
0.0067	24.0	3600	0.4972	0.0035	0	0.8569	0.7994	0.6973	0.5553	18.6165	0	0.9672	0.9447	0.9190	0.8885	3.7879
0.0067	25.0	3750	0.5022	0.0035	0	0.8529	0.7989	0.6962	0.5480	18.5148	0	0.9678	0.9454	0.9200	0.8903	3.7121
0.0067	26.0	3900	0.5223	0.0035	0	0.8509	0.7967	0.6938	0.5480	18.5148	0	0.9653	0.9426	0.9162	0.8850	3.9394
0.006	27.0	4050	0.5225	0.0035	0	0.8498	0.7977	0.6946	0.5482	18.4130	0	0.9671	0.9450	0.9194	0.8894	3.7879
0.006	28.0	4200	0.5249	0.0035	0	0.8539	0.8001	0.6969	0.5500	18.2096	0	0.9678	0.9449	0.9187	0.8877	3.7879
0.006	29.0	4350	0.5314	0.0035	0	0.8469	0.7906	0.6845	0.5375	18.9217	0	0.9678	0.9449	0.9187	0.8877	3.7879
0.0044	30.0	4500	0.5327	0.0035	0	0.8489	0.7929	0.6865	0.5401	18.8199	0	0.9678	0.9449	0.9187	0.8877	3.7879

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1