sentiment-lora-r8-1

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: 0.3074
• Accuracy: 0.8647
• Precision: 0.8398
• Recall: 0.8292
• F1: 0.8342

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
0.5627	1.0	122	0.5240	0.7168	0.6478	0.6246	0.6315
0.5032	2.0	244	0.5100	0.7393	0.6998	0.7256	0.7068
0.4595	3.0	366	0.4393	0.7845	0.7408	0.75	0.7450
0.4038	4.0	488	0.3939	0.8170	0.7898	0.7480	0.7634
0.3756	5.0	610	0.3823	0.8346	0.7990	0.8230	0.8086
0.3421	6.0	732	0.3498	0.8446	0.8126	0.8126	0.8126
0.3251	7.0	854	0.3386	0.8446	0.8158	0.8026	0.8086
0.3092	8.0	976	0.3341	0.8521	0.8284	0.8054	0.8154
0.2985	9.0	1098	0.3341	0.8546	0.8239	0.8272	0.8255
0.2941	10.0	1220	0.3237	0.8672	0.8449	0.8285	0.8360
0.2816	11.0	1342	0.3193	0.8672	0.8449	0.8285	0.8360
0.2765	12.0	1464	0.3179	0.8622	0.8361	0.8275	0.8316
0.2733	13.0	1586	0.3159	0.8697	0.8488	0.8303	0.8386
0.2702	14.0	1708	0.3141	0.8571	0.8281	0.8264	0.8272
0.2632	15.0	1830	0.3104	0.8622	0.8361	0.8275	0.8316
0.2556	16.0	1952	0.3081	0.8596	0.8325	0.8257	0.8290
0.2596	17.0	2074	0.3090	0.8697	0.8474	0.8328	0.8395
0.2576	18.0	2196	0.3081	0.8647	0.8377	0.8342	0.8359
0.2532	19.0	2318	0.3074	0.8647	0.8398	0.8292	0.8342
0.2444	20.0	2440	0.3074	0.8647	0.8398	0.8292	0.8342

Framework versions

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