README.md

metadata

library_name: transformers
license: mit
base_model: indobenchmark/indobert-base-p1
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: results
    results: []

results

This model is a fine-tuned version of indobenchmark/indobert-base-p1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5669
• Accuracy: 0.7895
• Precision: 0.8398
• Recall: 0.7895
• F1: 0.7991

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: 2.7820079535067715e-06
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 75
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.1527	1.0	6	1.1881	0.2895	0.3080	0.2895	0.1594
1.1409	2.0	12	1.1546	0.2895	0.3080	0.2895	0.1594
1.111	3.0	18	1.1169	0.2895	0.3099	0.2895	0.1618
1.1087	4.0	24	1.0870	0.3158	0.5620	0.3158	0.2119
1.104	5.0	30	1.0691	0.3421	0.6476	0.3421	0.2579
1.0784	6.0	36	1.0531	0.3684	0.6916	0.3684	0.3002
1.0681	7.0	42	1.0362	0.3947	0.7191	0.3947	0.3393
1.062	8.0	48	1.0119	0.5	0.7594	0.5	0.4877
1.0183	9.0	54	0.9743	0.4737	0.6711	0.4737	0.4801
1.0133	10.0	60	0.9400	0.5526	0.7075	0.5526	0.5677
0.9721	11.0	66	0.9150	0.6053	0.6900	0.6053	0.6217
0.9438	12.0	72	0.8792	0.6316	0.7042	0.6316	0.6474
0.9122	13.0	78	0.8155	0.6579	0.6856	0.6579	0.6666
0.8681	14.0	84	0.7988	0.6316	0.6707	0.6316	0.6424
0.8398	15.0	90	0.7718	0.6316	0.6917	0.6316	0.6478
0.8154	16.0	96	0.7375	0.6842	0.7206	0.6842	0.6942
0.7824	17.0	102	0.7162	0.7105	0.7372	0.7105	0.7154
0.7632	18.0	108	0.6953	0.6842	0.6981	0.6842	0.6862
0.7148	19.0	114	0.6705	0.6579	0.6917	0.6579	0.6701
0.7015	20.0	120	0.6466	0.6579	0.6917	0.6579	0.6701
0.6992	21.0	126	0.6408	0.6842	0.7303	0.6842	0.6981
0.6818	22.0	132	0.6199	0.7105	0.7432	0.7105	0.7215
0.6655	23.0	138	0.6283	0.7105	0.7738	0.7105	0.7218
0.6623	24.0	144	0.5984	0.7368	0.7586	0.7368	0.7430
0.615	25.0	150	0.5800	0.7632	0.7742	0.7632	0.7665
0.5923	26.0	156	0.5721	0.7632	0.7947	0.7632	0.7730
0.5976	27.0	162	0.5666	0.7895	0.8117	0.7895	0.7955
0.5631	28.0	168	0.5627	0.8158	0.8275	0.8158	0.8193
0.568	29.0	174	0.5747	0.7368	0.7853	0.7368	0.7469
0.5381	30.0	180	0.5548	0.7895	0.8117	0.7895	0.7955
0.5071	31.0	186	0.5673	0.7895	0.8398	0.7895	0.7991
0.5211	32.0	192	0.5915	0.7368	0.8211	0.7368	0.7485
0.4806	33.0	198	0.5635	0.8158	0.8514	0.8158	0.8237
0.4945	34.0	204	0.5626	0.8158	0.8514	0.8158	0.8237
0.4921	35.0	210	0.5669	0.7895	0.8398	0.7895	0.7991

Framework versions

• Transformers 4.56.2
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1