sentiment-pt-pl50-4

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: 0.3212
• Accuracy: 0.8847
• Precision: 0.8621
• Recall: 0.8584
• F1: 0.8602

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
0.5424	1.0	122	0.4889	0.7243	0.6616	0.6474	0.6529
0.4433	2.0	244	0.4064	0.8045	0.7900	0.7092	0.7306
0.3542	3.0	366	0.3433	0.8296	0.8066	0.7644	0.7803
0.3037	4.0	488	0.3535	0.8446	0.8193	0.7951	0.8055
0.2664	5.0	610	0.3154	0.8622	0.8361	0.8275	0.8316
0.2511	6.0	732	0.3063	0.8822	0.8656	0.8442	0.8537
0.234	7.0	854	0.3097	0.8797	0.8616	0.8424	0.8510
0.2232	8.0	976	0.3141	0.8647	0.8410	0.8267	0.8333
0.2074	9.0	1098	0.3008	0.8797	0.8651	0.8374	0.8494
0.1905	10.0	1220	0.2974	0.8697	0.8428	0.8428	0.8428
0.182	11.0	1342	0.2904	0.8722	0.8498	0.8371	0.8430
0.1757	12.0	1464	0.3064	0.8797	0.8651	0.8374	0.8494
0.1721	13.0	1586	0.3054	0.8772	0.8628	0.8331	0.8458
0.162	14.0	1708	0.3047	0.8822	0.8535	0.8717	0.8616
0.1521	15.0	1830	0.3181	0.8847	0.8581	0.8684	0.8629
0.1503	16.0	1952	0.3237	0.8872	0.8703	0.8527	0.8607
0.1473	17.0	2074	0.3206	0.8872	0.8634	0.8652	0.8643
0.1451	18.0	2196	0.3165	0.8847	0.8621	0.8584	0.8602
0.135	19.0	2318	0.3210	0.8872	0.8645	0.8627	0.8636
0.1373	20.0	2440	0.3212	0.8847	0.8621	0.8584	0.8602

Framework versions

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