Modelpublic

elfsmo/IndoBERT-SDGs-Oplib-Elsevier-Pruned

sourceHugging Facemitupdated 1y agoView on Hugging Face

0likes5downloads

Model Card

IndoBERT-SDGs-Oplib-Elsevier-Pruned

This model is a fine-tuned version of indobenchmark/indobert-base-p2 on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.1493
—Accuracy: 0.44
—F1 Micro: 0.8275
—F1 Macro: 0.7542
—Precision Micro: 0.8676
—Precision Macro: 0.8694
—Recall Micro: 0.7910
—Recall Macro: 0.7044
—Roc Auc: 0.8829
—Hamming Loss: 0.0568

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: 1.4287607820196737e-05
—trainbatchsize: 64
—evalbatchsize: 64
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—lrschedulerwarmup_ratio: 0.11880223991861419
—num_epochs: 8/20
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Micro	F1 Macro	Precision Micro	Precision Macro	Recall Micro	Recall Macro	Roc Auc	Hamming Loss
No log	1.0	179	0.3583	0.0955	0.4542	0.1318	0.6984	0.2599	0.3365	0.1234	0.6531	0.1393
No log	2.0	358	0.2449	0.2843	0.7145	0.4748	0.8380	0.6607	0.6227	0.3992	0.7988	0.0857
0.3435	3.0	537	0.1952	0.3787	0.7708	0.6324	0.8991	0.8543	0.6745	0.5373	0.8294	0.0691
0.3435	4.0	716	0.1742	0.4059	0.7996	0.7018	0.8838	0.8411	0.7301	0.6168	0.8550	0.0630
0.3435	5.0	895	0.1621	0.4149	0.8209	0.7444	0.8561	0.8088	0.7885	0.6950	0.8805	0.0593
0.1526	6.0	1074	0.1525	0.4432	0.8292	0.7615	0.8628	0.8685	0.7982	0.7135	0.8859	0.0566
0.1526	7.0	1253	0.1521	0.4363	0.8294	0.7629	0.8633	0.8591	0.7980	0.7098	0.8858	0.0566
0.1526	8.0	1432	0.1493	0.44	0.8275	0.7542	0.8676	0.8694	0.7910	0.7044	0.8829	0.0568

Framework versions

—Transformers 4.51.3
—Pytorch 2.7.0+cu126
—Datasets 3.6.0
—Tokenizers 0.21.1