sentiment-seq_bn-4

This model is a fine-tuned version of indolem/indobert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2917
• Accuracy: 0.8847
• Precision: 0.8648
• Recall: 0.8534
• F1: 0.8588

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: 5e-05
• train_batch_size: 30
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.5556	1.0	122	0.5148	0.7343	0.6852	0.6970	0.6899
0.476	2.0	244	0.4507	0.7870	0.7577	0.6918	0.7095
0.4238	3.0	366	0.4003	0.8195	0.7958	0.7473	0.7644
0.3735	4.0	488	0.3799	0.8371	0.8089	0.7872	0.7966
0.3548	5.0	610	0.3634	0.8471	0.8153	0.8168	0.8160
0.3213	6.0	732	0.3584	0.8421	0.8077	0.8208	0.8136
0.3085	7.0	854	0.3318	0.8622	0.8448	0.8125	0.8259
0.2981	8.0	976	0.3429	0.8672	0.8722	0.7985	0.8238
0.2788	9.0	1098	0.3304	0.8797	0.8795	0.8224	0.8439
0.259	10.0	1220	0.3076	0.8772	0.8535	0.8481	0.8507
0.2587	11.0	1342	0.3025	0.8747	0.8471	0.8538	0.8503
0.2391	12.0	1464	0.2990	0.8847	0.8697	0.8459	0.8565
0.2443	13.0	1586	0.2919	0.8797	0.8600	0.8449	0.8518
0.237	14.0	1708	0.3040	0.8772	0.8483	0.8631	0.8550
0.2176	15.0	1830	0.2937	0.8897	0.8649	0.8720	0.8683
0.2202	16.0	1952	0.2920	0.8822	0.8610	0.8517	0.8561
0.2203	17.0	2074	0.2923	0.8822	0.8585	0.8567	0.8575
0.2204	18.0	2196	0.2927	0.8847	0.8621	0.8584	0.8602
0.2124	19.0	2318	0.2920	0.8872	0.8672	0.8577	0.8622
0.2108	20.0	2440	0.2917	0.8847	0.8648	0.8534	0.8588

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1