ABL_trad_l

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: 3.2495
• Accuracy: 0.6833
• F1: 0.6809

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-06
• 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: 42

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.9228	1.0	2000	0.9053	0.5817	0.5807
0.8293	2.0	4000	0.8649	0.6	0.5927
0.7665	3.0	6000	0.8310	0.6217	0.6206
0.7292	4.0	8000	0.8270	0.6358	0.6316
0.6773	5.0	10000	0.8149	0.6558	0.6520
0.648	6.0	12000	0.8207	0.6492	0.6471
0.5912	7.0	14000	0.8353	0.6508	0.6487
0.5558	8.0	16000	0.8601	0.66	0.6585
0.5169	9.0	18000	0.9048	0.6617	0.6585
0.4678	10.0	20000	0.9497	0.6675	0.6646
0.4281	11.0	22000	1.0488	0.6633	0.6575
0.413	12.0	24000	1.1182	0.66	0.6557
0.389	13.0	26000	1.2184	0.6758	0.6718
0.3501	14.0	28000	1.3527	0.665	0.6613
0.3572	15.0	30000	1.4490	0.6692	0.6642
0.3136	16.0	32000	1.5910	0.6733	0.6713
0.3247	17.0	34000	1.7505	0.6708	0.6683
0.2824	18.0	36000	1.9347	0.6617	0.6551
0.2579	19.0	38000	2.0703	0.6733	0.6692
0.2641	20.0	40000	2.1537	0.6658	0.6609
0.1788	21.0	42000	2.2683	0.6758	0.6728
0.2099	22.0	44000	2.3347	0.6692	0.6670
0.1637	23.0	46000	2.4836	0.675	0.6712
0.1671	24.0	48000	2.5688	0.6775	0.6731
0.1455	25.0	50000	2.6975	0.6767	0.6699
0.1425	26.0	52000	2.7016	0.6742	0.6716
0.1406	27.0	54000	2.7527	0.6825	0.6785
0.1234	28.0	56000	2.8701	0.6758	0.6710
0.0967	29.0	58000	2.8947	0.685	0.6803
0.0864	30.0	60000	2.9296	0.6742	0.6723
0.0956	31.0	62000	2.9966	0.6808	0.6762
0.0835	32.0	64000	3.0406	0.6808	0.6759
0.073	33.0	66000	3.0750	0.6725	0.6680
0.0618	34.0	68000	3.0261	0.6808	0.6769
0.0833	35.0	70000	3.0812	0.685	0.6817
0.0478	36.0	72000	3.1352	0.6825	0.6784
0.0712	37.0	74000	3.1516	0.68	0.6780
0.0712	38.0	76000	3.2088	0.6708	0.6664
0.0407	39.0	78000	3.2520	0.6858	0.6828
0.0659	40.0	80000	3.2791	0.6792	0.6751
0.0468	41.0	82000	3.2433	0.6875	0.6826
0.0571	42.0	84000	3.2495	0.6833	0.6809

Framework versions

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