sentiment-pt-pl50-0

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: 0.2615
• Accuracy: 0.8947
• Precision: 0.8730
• Recall: 0.8730
• F1: 0.8730

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
0.5435	1.0	122	0.4679	0.7619	0.7114	0.6915	0.6994
0.4401	2.0	244	0.4174	0.8045	0.7728	0.8142	0.7832
0.3322	3.0	366	0.3163	0.8571	0.8419	0.8014	0.8175
0.2797	4.0	488	0.3073	0.8647	0.8424	0.8242	0.8324
0.2591	5.0	610	0.3091	0.8647	0.8329	0.8542	0.8420
0.247	6.0	732	0.2943	0.8747	0.8447	0.8638	0.8531
0.2379	7.0	854	0.3208	0.8722	0.8408	0.8746	0.8536
0.2119	8.0	976	0.2688	0.8897	0.8762	0.8520	0.8627
0.2031	9.0	1098	0.2807	0.8897	0.8632	0.8770	0.8695
0.192	10.0	1220	0.2733	0.8997	0.8988	0.8541	0.8723
0.195	11.0	1342	0.2573	0.9023	0.8814	0.8833	0.8824
0.1708	12.0	1464	0.2527	0.8997	0.8819	0.8741	0.8778
0.1642	13.0	1586	0.2566	0.9023	0.8913	0.8683	0.8786
0.1661	14.0	1708	0.2649	0.8972	0.8732	0.8823	0.8776
0.1624	15.0	1830	0.2631	0.9048	0.8816	0.8926	0.8868
0.1574	16.0	1952	0.2980	0.8747	0.8436	0.8763	0.8562
0.1496	17.0	2074	0.2547	0.8947	0.8730	0.8730	0.8730
0.1413	18.0	2196	0.2667	0.8947	0.8690	0.8830	0.8755
0.1478	19.0	2318	0.2663	0.8947	0.8708	0.8780	0.8743
0.1403	20.0	2440	0.2615	0.8947	0.8730	0.8730	0.8730

Framework versions

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