sentiment-pt-pl5-4

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: 0.3030
• Accuracy: 0.8847
• Precision: 0.8589
• Recall: 0.8659
• F1: 0.8623

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
0.5491	1.0	122	0.5238	0.7118	0.6569	0.6636	0.6598
0.467	2.0	244	0.4078	0.7895	0.7486	0.7685	0.7564
0.3738	3.0	366	0.3612	0.8496	0.8538	0.7711	0.7971
0.321	4.0	488	0.3049	0.8596	0.8406	0.8107	0.8233
0.2799	5.0	610	0.2869	0.8772	0.8693	0.8256	0.8431
0.2619	6.0	732	0.2957	0.8747	0.8523	0.8413	0.8465
0.2475	7.0	854	0.2894	0.8722	0.8528	0.8321	0.8413
0.2336	8.0	976	0.3088	0.8596	0.8374	0.8157	0.8253
0.2233	9.0	1098	0.2827	0.8772	0.8561	0.8431	0.8492
0.2053	10.0	1220	0.2771	0.8897	0.8659	0.8695	0.8676
0.1926	11.0	1342	0.2792	0.8847	0.8573	0.8709	0.8636
0.1837	12.0	1464	0.2857	0.8872	0.8687	0.8552	0.8615
0.1748	13.0	1586	0.2900	0.8922	0.8706	0.8687	0.8697
0.1664	14.0	1708	0.3100	0.8872	0.8587	0.8802	0.8681
0.1575	15.0	1830	0.3073	0.8872	0.8593	0.8777	0.8675
0.1553	16.0	1952	0.3023	0.8972	0.8781	0.8723	0.8751
0.1439	17.0	2074	0.3054	0.8847	0.8581	0.8684	0.8629
0.1509	18.0	2196	0.3081	0.8847	0.8599	0.8634	0.8616
0.1489	19.0	2318	0.3018	0.8897	0.8670	0.8670	0.8670
0.146	20.0	2440	0.3030	0.8847	0.8589	0.8659	0.8623

Framework versions

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