sentiment-pt-pl20-2

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: 0.3072
• Accuracy: 0.8872
• Precision: 0.8634
• Recall: 0.8652
• F1: 0.8643

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
0.5414	1.0	122	0.5076	0.7118	0.6352	0.5986	0.6042
0.4441	2.0	244	0.4079	0.7945	0.7579	0.7896	0.7679
0.3541	3.0	366	0.3338	0.8496	0.8289	0.7961	0.8096
0.3048	4.0	488	0.3370	0.8596	0.8316	0.8282	0.8298
0.2849	5.0	610	0.3066	0.8622	0.8373	0.8250	0.8307
0.2581	6.0	732	0.2884	0.8772	0.8547	0.8456	0.8500
0.2486	7.0	854	0.2819	0.8797	0.8539	0.8574	0.8556
0.2303	8.0	976	0.2881	0.8822	0.8715	0.8367	0.8513
0.2167	9.0	1098	0.2772	0.8822	0.8639	0.8467	0.8545
0.1948	10.0	1220	0.3049	0.8822	0.8674	0.8417	0.8529
0.1901	11.0	1342	0.2994	0.8847	0.8697	0.8459	0.8565
0.1911	12.0	1464	0.2923	0.8847	0.8648	0.8534	0.8588
0.1683	13.0	1586	0.2878	0.8922	0.8719	0.8662	0.8690
0.1711	14.0	1708	0.3060	0.8897	0.8618	0.8820	0.8707
0.1627	15.0	1830	0.2843	0.8997	0.8778	0.8816	0.8797
0.1561	16.0	1952	0.2887	0.8922	0.8694	0.8712	0.8703
0.147	17.0	2074	0.3024	0.8947	0.8730	0.8730	0.8730
0.1358	18.0	2196	0.3073	0.8897	0.8659	0.8695	0.8676
0.1506	19.0	2318	0.3115	0.8897	0.8695	0.8620	0.8656
0.151	20.0	2440	0.3072	0.8872	0.8634	0.8652	0.8643

Framework versions

• Transformers 4.39.3
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.15.2