FNST_trad_h

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: 1.3344
• Accuracy: 0.6633
• F1: 0.6496

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: 12
• eval_batch_size: 12
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 24

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.1493	1.0	1000	1.1016	0.5125	0.3623
0.9936	2.0	2000	0.9526	0.5942	0.5059
0.9056	3.0	3000	0.8908	0.6308	0.5971
0.8394	4.0	4000	0.8537	0.6425	0.6247
0.7877	5.0	5000	0.8308	0.6525	0.6362
0.7706	6.0	6000	0.8333	0.6608	0.6435
0.7142	7.0	7000	0.8258	0.6625	0.6505
0.6946	8.0	8000	0.8234	0.6617	0.6444
0.6525	9.0	9000	0.8367	0.6625	0.6518
0.6277	10.0	10000	0.8403	0.6583	0.6498
0.594	11.0	11000	0.8604	0.6583	0.6492
0.5691	12.0	12000	0.8701	0.665	0.6572
0.5268	13.0	13000	0.8956	0.6617	0.6534
0.5042	14.0	14000	0.9284	0.6583	0.6493
0.48	15.0	15000	0.9583	0.6567	0.6454
0.4507	16.0	16000	0.9734	0.6592	0.6479
0.4073	17.0	17000	1.0266	0.6617	0.6458
0.3838	18.0	18000	1.0626	0.6625	0.6477
0.3668	19.0	19000	1.0924	0.6575	0.6413
0.3439	20.0	20000	1.1276	0.6692	0.6595
0.3186	21.0	21000	1.1924	0.6608	0.6405
0.2926	22.0	22000	1.2392	0.6617	0.6468
0.2767	23.0	23000	1.2820	0.6658	0.6493
0.2531	24.0	24000	1.3344	0.6633	0.6496

Framework versions

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