sentiment-pt-pl30-3

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: 0.3054
• Accuracy: 0.8872
• Precision: 0.8614
• Recall: 0.8702
• F1: 0.8656

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
0.5456	1.0	122	0.4916	0.7469	0.6913	0.6534	0.6642
0.4369	2.0	244	0.4108	0.8120	0.7803	0.8220	0.7912
0.3316	3.0	366	0.3294	0.8571	0.8463	0.7964	0.8152
0.2909	4.0	488	0.3019	0.8772	0.8547	0.8456	0.8500
0.2584	5.0	610	0.3023	0.8697	0.8428	0.8428	0.8428
0.237	6.0	732	0.3020	0.8647	0.8359	0.8392	0.8376
0.2186	7.0	854	0.2989	0.8722	0.8425	0.8571	0.8491
0.2108	8.0	976	0.2961	0.8872	0.8687	0.8552	0.8615
0.1898	9.0	1098	0.3013	0.8747	0.8499	0.8463	0.8481
0.1894	10.0	1220	0.3231	0.8747	0.8537	0.8388	0.8457
0.1817	11.0	1342	0.3012	0.8772	0.8524	0.8506	0.8515
0.1723	12.0	1464	0.2979	0.8647	0.8377	0.8342	0.8359
0.1547	13.0	1586	0.2937	0.8697	0.8449	0.8378	0.8412
0.1569	14.0	1708	0.3065	0.8697	0.8383	0.8628	0.8486
0.1442	15.0	1830	0.2884	0.8772	0.8504	0.8556	0.8530
0.1435	16.0	1952	0.3016	0.8797	0.8530	0.8599	0.8563
0.1378	17.0	2074	0.3114	0.8772	0.8478	0.8656	0.8557
0.1377	18.0	2196	0.3096	0.8772	0.8483	0.8631	0.8550
0.1307	19.0	2318	0.3065	0.8797	0.8539	0.8574	0.8556
0.126	20.0	2440	0.3054	0.8872	0.8614	0.8702	0.8656

Framework versions

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