sentiment-lora-r8-4

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: 0.2847
• Accuracy: 0.8672
• Precision: 0.8423
• Recall: 0.8335
• F1: 0.8377

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
0.5577	1.0	122	0.5372	0.7193	0.6692	0.6814	0.6738
0.5012	2.0	244	0.4771	0.7669	0.7370	0.6501	0.6651
0.4626	3.0	366	0.4250	0.8070	0.7756	0.7360	0.7504
0.4055	4.0	488	0.3896	0.8346	0.7996	0.8055	0.8024
0.3709	5.0	610	0.3578	0.8296	0.7961	0.7869	0.7912
0.3385	6.0	732	0.3523	0.8371	0.8017	0.8172	0.8086
0.3276	7.0	854	0.3307	0.8521	0.8271	0.8079	0.8164
0.3133	8.0	976	0.3256	0.8571	0.8381	0.8064	0.8196
0.3039	9.0	1098	0.3282	0.8647	0.8491	0.8142	0.8286
0.2831	10.0	1220	0.3142	0.8596	0.8316	0.8282	0.8298
0.2798	11.0	1342	0.3034	0.8747	0.8523	0.8413	0.8465
0.269	12.0	1464	0.3002	0.8672	0.8479	0.8235	0.8342
0.2699	13.0	1586	0.2973	0.8697	0.8428	0.8428	0.8428
0.2657	14.0	1708	0.2985	0.8722	0.8445	0.8496	0.8470
0.2537	15.0	1830	0.2886	0.8672	0.8423	0.8335	0.8377
0.2529	16.0	1952	0.2878	0.8647	0.8387	0.8317	0.8351
0.2565	17.0	2074	0.2877	0.8722	0.8463	0.8446	0.8454
0.2514	18.0	2196	0.2857	0.8672	0.8412	0.8360	0.8385
0.2517	19.0	2318	0.2844	0.8697	0.8438	0.8403	0.8420
0.2512	20.0	2440	0.2847	0.8672	0.8423	0.8335	0.8377

Framework versions

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