robertuito-esp

This model is a fine-tuned version of pysentimiento/robertuito-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• F1: 0.8528
• Loss: 0.5317

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: 2.728093668459819e-05
• train_batch_size: 4
• eval_batch_size: 4
• 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: 3
• mixed_precision_training: Native AMP
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	F1	Validation Loss
0.6839	0.0313	50	0.7302	0.5988
0.5925	0.0626	100	0.7898	0.4998
0.5793	0.0939	150	0.8199	0.4645
0.4964	0.1252	200	0.7882	0.5048
0.4979	0.1565	250	0.8483	0.4355
0.5462	0.1879	300	0.8589	0.4264
0.4271	0.2192	350	0.8325	0.4819
0.4801	0.2505	400	0.8688	0.4230
0.4246	0.2818	450	0.8731	0.4355
0.4435	0.3131	500	0.8730	0.4197
0.3257	0.3444	550	0.8710	0.4488
0.4379	0.3757	600	0.8652	0.4428
0.4813	0.4070	650	0.8647	0.4094
0.49	0.4383	700	0.8263	0.4768
0.3999	0.4696	750	0.8467	0.4463
0.3629	0.5009	800	0.8523	0.4403
0.403	0.5322	850	0.8670	0.4416
0.3329	0.5636	900	0.8547	0.4821
0.4652	0.5949	950	0.8509	0.4877
0.4348	0.6262	1000	0.8565	0.4801
0.3317	0.6575	1050	0.8423	0.4966
0.46	0.6888	1100	0.8327	0.5275
0.3597	0.7201	1150	0.8508	0.5103
0.358	0.7514	1200	0.8522	0.4784
0.3856	0.7827	1250	0.8693	0.4824
0.3635	0.8140	1300	0.8729	0.4662
0.4228	0.8453	1350	0.8612	0.4617
0.3565	0.8766	1400	0.8628	0.4627
0.3035	0.9080	1450	0.8672	0.4734
0.407	0.9393	1500	0.8641	0.4566
0.3273	0.9706	1550	0.8531	0.4912
0.2871	1.0019	1600	0.8673	0.4843
0.2829	1.0332	1650	0.8591	0.4843
0.2512	1.0645	1700	0.8588	0.5057
0.2945	1.0958	1750	0.8448	0.5404
0.3107	1.1271	1800	0.8647	0.4773
0.2441	1.1584	1850	0.8530	0.5198
0.2744	1.1897	1900	0.8669	0.5051
0.2469	1.2210	1950	0.8569	0.5106
0.2532	1.2523	2000	0.8692	0.5018
0.2995	1.2837	2050	0.8651	0.5020
0.2461	1.3150	2100	0.8571	0.5256
0.2463	1.3463	2150	0.8653	0.5064
0.257	1.3776	2200	0.8669	0.4898
0.2294	1.4089	2250	0.8673	0.4992
0.2621	1.4402	2300	0.8652	0.5104
0.2373	1.4715	2350	0.8487	0.5130
0.2367	1.5028	2400	0.8448	0.5559
0.2464	1.5341	2450	0.8653	0.5204
0.2348	1.5654	2500	0.8693	0.5159
0.2069	1.5967	2550	0.8588	0.5004
0.2213	1.6281	2600	0.8592	0.5359
0.2264	1.6594	2650	0.8652	0.5244
0.2296	1.6907	2700	0.8611	0.5211
0.2366	1.7220	2750	0.8592	0.5117
0.2392	1.7533	2800	0.8706	0.4882
0.2636	1.7846	2850	0.8713	0.4988
0.2426	1.8159	2900	0.8732	0.4955
0.2541	1.8472	2950	0.8690	0.4957
0.2625	1.8785	3000	0.8752	0.4843
0.2151	1.9098	3050	0.8710	0.5104
0.2214	1.9411	3100	0.8710	0.5103
0.2708	1.9724	3150	0.8639	0.4959
0.2593	2.0038	3200	0.8652	0.5207
0.2233	2.0351	3250	0.8611	0.5260
0.2223	2.0664	3300	0.8671	0.5186
0.2262	2.0977	3350	0.8705	0.4925
0.2297	2.1290	3400	0.8610	0.5214
0.2042	2.1603	3450	0.8590	0.5329
0.2238	2.1916	3500	0.8489	0.5318
0.2109	2.2229	3550	0.8570	0.5286
0.226	2.2542	3600	0.8630	0.5232
0.2594	2.2855	3650	0.8651	0.5132
0.2264	2.3168	3700	0.8570	0.5239
0.2025	2.3482	3750	0.8590	0.5274
0.2064	2.3795	3800	0.8609	0.5045
0.2004	2.4108	3850	0.8650	0.5114
0.2278	2.4421	3900	0.8489	0.5277
0.2193	2.4734	3950	0.8610	0.5227
0.2231	2.5047	4000	0.8609	0.5109
0.207	2.5360	4050	0.8566	0.5087
0.1995	2.5673	4100	0.8630	0.5221
0.2125	2.5986	4150	0.8610	0.5242
0.2014	2.6299	4200	0.8550	0.5371
0.2118	2.6612	4250	0.8591	0.5321
0.1995	2.6925	4300	0.8550	0.5375
0.2258	2.7239	4350	0.8550	0.5352
0.1994	2.7552	4400	0.8570	0.5390
0.2235	2.7865	4450	0.8570	0.5306
0.2109	2.8178	4500	0.8530	0.5452
0.2091	2.8491	4550	0.8550	0.5345
0.1994	2.8804	4600	0.8550	0.5356
0.2134	2.9117	4650	0.8529	0.5368
0.2111	2.9430	4700	0.8508	0.5317
0.1995	2.9743	4750	0.8528	0.5317

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.5.0
• Tokenizers 0.21.1