sentiment-lora-r4a2d0.1-0

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: 0.3483
• Accuracy: 0.8446
• Precision: 0.8111
• Recall: 0.8201
• F1: 0.8153

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
0.5617	1.0	122	0.5117	0.7193	0.6580	0.6514	0.6543
0.5046	2.0	244	0.4917	0.7419	0.7042	0.7324	0.7112
0.4798	3.0	366	0.4466	0.7594	0.7129	0.7248	0.7179
0.4374	4.0	488	0.3994	0.8195	0.7866	0.7648	0.7741
0.4037	5.0	610	0.4150	0.7845	0.7480	0.7800	0.7575
0.3741	6.0	732	0.3737	0.8371	0.8028	0.8072	0.8049
0.3574	7.0	854	0.3776	0.8221	0.7845	0.7991	0.7909
0.3387	8.0	976	0.3654	0.8446	0.8120	0.8151	0.8135
0.3293	9.0	1098	0.3627	0.8371	0.8021	0.8122	0.8068
0.3209	10.0	1220	0.3553	0.8371	0.8032	0.8047	0.8040
0.2967	11.0	1342	0.3674	0.8346	0.7989	0.8130	0.8052
0.2928	12.0	1464	0.3707	0.8321	0.7960	0.8112	0.8027
0.2967	13.0	1586	0.3514	0.8471	0.8153	0.8168	0.8160
0.2934	14.0	1708	0.3507	0.8421	0.8083	0.8158	0.8119
0.2811	15.0	1830	0.3553	0.8346	0.7991	0.8105	0.8043
0.2738	16.0	1952	0.3555	0.8421	0.8077	0.8208	0.8136
0.2717	17.0	2074	0.3468	0.8496	0.8174	0.8236	0.8204
0.278	18.0	2196	0.3510	0.8421	0.8080	0.8183	0.8127
0.2701	19.0	2318	0.3471	0.8471	0.8142	0.8218	0.8178
0.2722	20.0	2440	0.3483	0.8446	0.8111	0.8201	0.8153

Framework versions

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