sentiment-lora-r16a1d0.1-0

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: 0.2912
• Accuracy: 0.8647
• Precision: 0.8346
• Recall: 0.8442
• F1: 0.8391

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
0.559	1.0	122	0.5023	0.7268	0.6671	0.6592	0.6627
0.4818	2.0	244	0.4518	0.7569	0.7248	0.7605	0.7318
0.4129	3.0	366	0.3967	0.8246	0.7876	0.7959	0.7914
0.3519	4.0	488	0.3626	0.8496	0.8193	0.8161	0.8177
0.3191	5.0	610	0.3720	0.8471	0.8130	0.8393	0.8235
0.2977	6.0	732	0.3482	0.8546	0.8216	0.8422	0.8303
0.2861	7.0	854	0.3343	0.8672	0.8363	0.8535	0.8439
0.2662	8.0	976	0.3213	0.8622	0.8314	0.8425	0.8365
0.2618	9.0	1098	0.3246	0.8697	0.8399	0.8528	0.8458
0.2556	10.0	1220	0.3065	0.8596	0.8307	0.8307	0.8307
0.2353	11.0	1342	0.3148	0.8722	0.8416	0.8621	0.8505
0.24	12.0	1464	0.3098	0.8747	0.8451	0.8613	0.8524
0.2346	13.0	1586	0.2989	0.8772	0.8524	0.8506	0.8515
0.2367	14.0	1708	0.3001	0.8697	0.8399	0.8528	0.8458
0.2248	15.0	1830	0.3040	0.8722	0.8420	0.8596	0.8498
0.2174	16.0	1952	0.3016	0.8697	0.8386	0.8603	0.8479
0.2112	17.0	2074	0.2887	0.8647	0.8346	0.8442	0.8391
0.2162	18.0	2196	0.2980	0.8722	0.8416	0.8621	0.8505
0.2124	19.0	2318	0.2892	0.8697	0.8411	0.8478	0.8443
0.2118	20.0	2440	0.2912	0.8647	0.8346	0.8442	0.8391

Framework versions

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