sentiment-pt-pl50-1

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: 0.3334
• Accuracy: 0.8822
• Precision: 0.8585
• Recall: 0.8567
• F1: 0.8575

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
0.5457	1.0	122	0.4914	0.7419	0.6837	0.6473	0.6575
0.4289	2.0	244	0.3963	0.8296	0.8009	0.8494	0.8123
0.3245	3.0	366	0.3338	0.8521	0.8496	0.7804	0.8038
0.2774	4.0	488	0.3109	0.8571	0.8365	0.8089	0.8206
0.2577	5.0	610	0.2983	0.8747	0.8488	0.8488	0.8488
0.2471	6.0	732	0.2961	0.8672	0.8372	0.8485	0.8425
0.2405	7.0	854	0.2966	0.8722	0.8420	0.8596	0.8498
0.2135	8.0	976	0.2851	0.8747	0.8605	0.8288	0.8422
0.2057	9.0	1098	0.2847	0.8797	0.8600	0.8449	0.8518
0.1962	10.0	1220	0.2998	0.8797	0.8715	0.8299	0.8468
0.1875	11.0	1342	0.2826	0.8772	0.8547	0.8456	0.8500
0.1715	12.0	1464	0.2922	0.8797	0.8539	0.8574	0.8556
0.1697	13.0	1586	0.3065	0.8822	0.8624	0.8492	0.8553
0.157	14.0	1708	0.3172	0.8847	0.8556	0.8784	0.8654
0.1502	15.0	1830	0.3085	0.8822	0.8555	0.8642	0.8596
0.1474	16.0	1952	0.3126	0.8747	0.8457	0.8588	0.8517
0.1405	17.0	2074	0.3143	0.8822	0.8585	0.8567	0.8575
0.1283	18.0	2196	0.3295	0.8772	0.8496	0.8581	0.8537
0.1402	19.0	2318	0.3333	0.8822	0.8585	0.8567	0.8575
0.1332	20.0	2440	0.3334	0.8822	0.8585	0.8567	0.8575

Framework versions

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