sentiment-unipelt-3

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: 0.3098
• Accuracy: 0.8797
• Precision: 0.8539
• Recall: 0.8574
• F1: 0.8556

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
0.5539	1.0	122	0.4987	0.7444	0.6873	0.6541	0.6641
0.4551	2.0	244	0.4491	0.7820	0.7544	0.7982	0.7618
0.3708	3.0	366	0.3649	0.8271	0.8039	0.7601	0.7764
0.3145	4.0	488	0.3169	0.8747	0.8671	0.8213	0.8394
0.2649	5.0	610	0.3174	0.8471	0.8135	0.8268	0.8195
0.252	6.0	732	0.2997	0.8672	0.8354	0.8610	0.8460
0.2408	7.0	854	0.2899	0.8622	0.8298	0.8575	0.8408
0.2192	8.0	976	0.2855	0.8647	0.8359	0.8392	0.8376
0.2024	9.0	1098	0.2897	0.8596	0.8293	0.8357	0.8324
0.1944	10.0	1220	0.2704	0.8672	0.8423	0.8335	0.8377
0.1812	11.0	1342	0.2703	0.8722	0.8431	0.8546	0.8484
0.1707	12.0	1464	0.2839	0.8772	0.8483	0.8631	0.8550
0.1619	13.0	1586	0.2882	0.8747	0.8479	0.8513	0.8496
0.149	14.0	1708	0.3145	0.8747	0.8436	0.8738	0.8556
0.1503	15.0	1830	0.2928	0.8847	0.8609	0.8609	0.8609
0.1415	16.0	1952	0.3014	0.8822	0.8656	0.8442	0.8537
0.1327	17.0	2074	0.3084	0.8822	0.8564	0.8617	0.8590
0.1343	18.0	2196	0.3049	0.8722	0.8425	0.8571	0.8491
0.1145	19.0	2318	0.3094	0.8797	0.8539	0.8574	0.8556
0.1203	20.0	2440	0.3098	0.8797	0.8539	0.8574	0.8556

Framework versions

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