sentiment-analysis

This model is a fine-tuned version of mdhugol/indonesia-bert-sentiment-classification on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7071
• Accuracy: 0.7689

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: 1e-08
• train_batch_size: 8
• eval_batch_size: 8
• seed: 41
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.9403	1.7986	500	3.9602	0.1973
3.5081	3.5971	1000	3.4530	0.2270
2.974	5.3957	1500	2.9278	0.2622
2.4812	7.1942	2000	2.4365	0.3176
2.065	8.9928	2500	2.0129	0.3703
1.7003	10.7914	3000	1.6758	0.4703
1.426	12.5899	3500	1.4297	0.5378
1.2629	14.3885	4000	1.2590	0.5919
1.1323	16.1871	4500	1.1440	0.6122
1.042	17.9856	5000	1.0668	0.6338
0.9725	19.7842	5500	1.0117	0.6662
0.9322	21.5827	6000	0.9717	0.6824
0.8992	23.3813	6500	0.9395	0.6959
0.8896	25.1799	7000	0.9145	0.7122
0.8666	26.9784	7500	0.8927	0.7203
0.8441	28.7770	8000	0.8739	0.7270
0.8037	30.5755	8500	0.8582	0.7365
0.7941	32.3741	9000	0.8438	0.7432
0.7893	34.1727	9500	0.8309	0.7392
0.7715	35.9712	10000	0.8191	0.7392
0.7444	37.7698	10500	0.8092	0.7405
0.771	39.5683	11000	0.8005	0.7405
0.7395	41.3669	11500	0.7919	0.7459
0.7557	43.1655	12000	0.7840	0.7486
0.7207	44.9640	12500	0.7771	0.75
0.7245	46.7626	13000	0.7705	0.7527
0.7135	48.5612	13500	0.7647	0.7541
0.7336	50.3597	14000	0.7591	0.7541
0.6999	52.1583	14500	0.7541	0.7554
0.715	53.9568	15000	0.7493	0.7568
0.6974	55.7554	15500	0.7450	0.7581
0.6847	57.5540	16000	0.7408	0.7581
0.7009	59.3525	16500	0.7372	0.7595
0.6781	61.1511	17000	0.7338	0.7608
0.6874	62.9496	17500	0.7305	0.7622
0.6861	64.7482	18000	0.7275	0.7622
0.6848	66.5468	18500	0.7249	0.7635
0.6617	68.3453	19000	0.7228	0.7649
0.6845	70.1439	19500	0.7204	0.7662
0.6619	71.9424	20000	0.7183	0.7662
0.6681	73.7410	20500	0.7165	0.7662
0.6792	75.5396	21000	0.7148	0.7662
0.6687	77.3381	21500	0.7133	0.7676
0.6779	79.1367	22000	0.7120	0.7676
0.6679	80.9353	22500	0.7109	0.7689
0.6531	82.7338	23000	0.7101	0.7689
0.6592	84.5324	23500	0.7092	0.7689
0.6578	86.3309	24000	0.7086	0.7689
0.6552	88.1295	24500	0.7081	0.7689
0.6738	89.9281	25000	0.7077	0.7689
0.6517	91.7266	25500	0.7074	0.7689
0.6694	93.5252	26000	0.7073	0.7689
0.6543	95.3237	26500	0.7072	0.7689
0.6601	97.1223	27000	0.7071	0.7689
0.6524	98.9209	27500	0.7071	0.7689

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.1