sentiment-lora-r4a2d0.05-0

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: 0.3486
• Accuracy: 0.8396
• Precision: 0.8055
• Recall: 0.8115
• F1: 0.8084

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
0.5619	1.0	122	0.5127	0.7168	0.6536	0.6446	0.6484
0.5059	2.0	244	0.4967	0.7343	0.6956	0.7220	0.7022
0.4822	3.0	366	0.4506	0.7469	0.7006	0.7159	0.7065
0.4402	4.0	488	0.3984	0.8195	0.7876	0.7623	0.7728
0.4068	5.0	610	0.4136	0.7870	0.7473	0.7718	0.7561
0.3791	6.0	732	0.3771	0.8321	0.7972	0.7987	0.7979
0.3635	7.0	854	0.3916	0.8195	0.7822	0.8048	0.7912
0.3433	8.0	976	0.3799	0.8296	0.7934	0.8019	0.7974
0.3379	9.0	1098	0.3714	0.8271	0.7903	0.8026	0.7959
0.3296	10.0	1220	0.3635	0.8371	0.8032	0.8047	0.8040
0.3105	11.0	1342	0.3652	0.8296	0.7933	0.8044	0.7984
0.3024	12.0	1464	0.3702	0.8346	0.7988	0.8180	0.8069
0.309	13.0	1586	0.3512	0.8371	0.8032	0.8047	0.8040
0.3021	14.0	1708	0.3505	0.8396	0.8060	0.8090	0.8075
0.2903	15.0	1830	0.3553	0.8421	0.8077	0.8208	0.8136
0.2834	16.0	1952	0.3530	0.8396	0.8046	0.8215	0.8119
0.2811	17.0	2074	0.3471	0.8446	0.8120	0.8151	0.8135
0.288	18.0	2196	0.3505	0.8446	0.8107	0.8226	0.8161
0.277	19.0	2318	0.3479	0.8396	0.8055	0.8115	0.8084
0.2775	20.0	2440	0.3486	0.8396	0.8055	0.8115	0.8084

Framework versions

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