spa-eng-pos-tagging-v2.1

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

• Loss: 0.1523
• Accuracy: 0.9516
• Precision: 0.9445
• Recall: 0.8846
• F1: 0.8831
• Hamming Loss: 0.0484

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 14

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Hamming Loss
0.3806	1.0	1744	0.2730	0.8940	0.8903	0.8246	0.8252	0.1060
0.2615	2.0	3488	0.2134	0.9181	0.9058	0.8514	0.8464	0.0819
0.1864	3.0	5232	0.1895	0.9284	0.9224	0.8626	0.8608	0.0716
0.1721	4.0	6976	0.1769	0.9352	0.9287	0.8676	0.8667	0.0648
0.1462	5.0	8720	0.1840	0.9355	0.9310	0.8663	0.8670	0.0645
0.1175	6.0	10464	0.1553	0.9487	0.9399	0.8820	0.8794	0.0513
0.0949	7.0	12208	0.1523	0.9516	0.9445	0.8846	0.8831	0.0484
0.0833	8.0	13952	0.1564	0.9535	0.9439	0.8881	0.8847	0.0465
0.0722	9.0	15696	0.1566	0.9549	0.9472	0.8890	0.8867	0.0451
0.0571	10.0	17440	0.1649	0.9565	0.9513	0.8875	0.8880	0.0435
0.0516	11.0	19184	0.1669	0.9587	0.9523	0.8913	0.8905	0.0413
0.0392	12.0	20928	0.1711	0.9598	0.9524	0.8925	0.8911	0.0402
0.03	13.0	22672	0.1800	0.9603	0.9533	0.8935	0.8921	0.0397
0.0308	14.0	24416	0.1827	0.9607	0.9537	0.8938	0.8924	0.0393

Framework versions

• Transformers 4.32.0
• Pytorch 2.0.1+cu118
• Tokenizers 0.13.3