sentiment-lora-r2a2d0.05-1

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: 0.3638
• Accuracy: 0.8446
• Precision: 0.8193
• Recall: 0.7951
• F1: 0.8055

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
0.5663	1.0	122	0.5216	0.7293	0.6677	0.6510	0.6572
0.5149	2.0	244	0.5134	0.7243	0.6758	0.6899	0.6810
0.4925	3.0	366	0.4821	0.7569	0.7055	0.6980	0.7014
0.4608	4.0	488	0.4654	0.7644	0.7150	0.7083	0.7114
0.4493	5.0	610	0.4600	0.7569	0.7126	0.7305	0.7193
0.4257	6.0	732	0.4307	0.7870	0.7433	0.7318	0.7369
0.4178	7.0	854	0.4181	0.7970	0.7552	0.7614	0.7581
0.3977	8.0	976	0.3972	0.8070	0.7687	0.7560	0.7617
0.3946	9.0	1098	0.3937	0.8145	0.7779	0.7663	0.7716
0.3762	10.0	1220	0.3874	0.8246	0.7995	0.7584	0.7738
0.3727	11.0	1342	0.3787	0.8321	0.8014	0.7837	0.7915
0.3626	12.0	1464	0.3750	0.8371	0.8059	0.7947	0.7999
0.359	13.0	1586	0.3728	0.8296	0.8066	0.7644	0.7803
0.3488	14.0	1708	0.3709	0.8296	0.8049	0.7669	0.7816
0.3445	15.0	1830	0.3667	0.8421	0.8131	0.7983	0.8050
0.3344	16.0	1952	0.3656	0.8421	0.8142	0.7958	0.8040
0.3339	17.0	2074	0.3654	0.8396	0.8128	0.7890	0.7992
0.3357	18.0	2196	0.3638	0.8421	0.8154	0.7933	0.8029
0.3357	19.0	2318	0.3646	0.8421	0.8154	0.7933	0.8029
0.3359	20.0	2440	0.3638	0.8446	0.8193	0.7951	0.8055

Framework versions

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