goemotions_bertspanish_finetunig_g

This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.2987
• Accuracy: 0.465
• F1: 0.3459

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: 1e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 64

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
2.6699	1.0	2000	2.6095	0.3042	0.0253
2.501	2.0	4000	2.3754	0.3925	0.0803
2.3358	3.0	6000	2.2518	0.4217	0.0978
2.2927	4.0	8000	2.1618	0.4425	0.1188
2.12	5.0	10000	2.0765	0.4575	0.1421
2.0914	6.0	12000	2.0328	0.4583	0.1731
2.0042	7.0	14000	1.9635	0.4783	0.2000
1.9661	8.0	16000	1.9428	0.4792	0.2261
1.8821	9.0	18000	1.9069	0.485	0.2251
1.8648	10.0	20000	1.8867	0.4917	0.2451
1.7464	11.0	22000	1.8609	0.5017	0.2543
1.763	12.0	24000	1.8369	0.5025	0.2698
1.7309	13.0	26000	1.8461	0.4967	0.2780
1.648	14.0	28000	1.8432	0.5017	0.2783
1.6077	15.0	30000	1.8419	0.4892	0.2911
1.5947	16.0	32000	1.8492	0.4983	0.2904
1.5088	17.0	34000	1.8646	0.5125	0.2927
1.5087	18.0	36000	1.8509	0.5025	0.2987
1.4678	19.0	38000	1.8596	0.4958	0.3105
1.3912	20.0	40000	1.8848	0.5025	0.3018
1.3895	21.0	42000	1.8688	0.4883	0.2993
1.3499	22.0	44000	1.9051	0.4992	0.3124
1.3294	23.0	46000	1.9304	0.4942	0.3258
1.2276	24.0	48000	1.9361	0.4933	0.3374
1.2186	25.0	50000	1.9587	0.4925	0.3423
1.2256	26.0	52000	1.9841	0.5	0.3446
1.1503	27.0	54000	2.0229	0.4842	0.3446
1.1663	28.0	56000	2.0386	0.4817	0.3409
1.0824	29.0	58000	2.0353	0.4892	0.3388
1.0552	30.0	60000	2.0629	0.5058	0.3530
1.0633	31.0	62000	2.0942	0.4883	0.3559
0.9907	32.0	64000	2.1115	0.49	0.3479
0.9964	33.0	66000	2.1404	0.4867	0.3523
0.9144	34.0	68000	2.1559	0.4908	0.3476
0.9246	35.0	70000	2.2217	0.475	0.3360
0.8605	36.0	72000	2.2469	0.475	0.3426
0.8902	37.0	74000	2.2601	0.4908	0.3459
0.8197	38.0	76000	2.3256	0.4675	0.3420
0.776	39.0	78000	2.3163	0.4925	0.3483
0.8043	40.0	80000	2.3630	0.4808	0.3499
0.7914	41.0	82000	2.4009	0.475	0.3457
0.7128	42.0	84000	2.4279	0.4767	0.3510
0.7011	43.0	86000	2.4393	0.4758	0.3310
0.6647	44.0	88000	2.4876	0.4808	0.3455
0.6663	45.0	90000	2.5062	0.4858	0.3584
0.6134	46.0	92000	2.5754	0.4758	0.3476
0.6141	47.0	94000	2.5886	0.4775	0.3575
0.6337	48.0	96000	2.6593	0.4675	0.3433
0.6029	49.0	98000	2.6852	0.4658	0.3425
0.5546	50.0	100000	2.6974	0.4708	0.3488
0.5331	51.0	102000	2.7254	0.4708	0.3418
0.5025	52.0	104000	2.7650	0.4742	0.3497
0.4917	53.0	106000	2.7831	0.4767	0.3425
0.4526	54.0	108000	2.8553	0.4667	0.3489
0.4399	55.0	110000	2.9210	0.4667	0.3539
0.4761	56.0	112000	2.9234	0.4717	0.3542
0.3981	57.0	114000	2.9558	0.4708	0.3489
0.3861	58.0	116000	3.0372	0.4725	0.3522
0.4151	59.0	118000	3.0529	0.4683	0.3429
0.3756	60.0	120000	3.1011	0.4725	0.3519
0.3688	61.0	122000	3.1638	0.4667	0.3469
0.3301	62.0	124000	3.2081	0.4708	0.3511
0.3236	63.0	126000	3.2128	0.4733	0.3574
0.3212	64.0	128000	3.2987	0.465	0.3459

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1