klasifikasiburung_new1

This model is a fine-tuned version of RobertZ2011/resnet-18-birb on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3710
• Accuracy: 0.7475
• Precision: 0.7511
• Recall: 0.7475
• F1: 0.7421

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-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
2.5853	1.0	94	2.3846	0.5300	0.6277	0.5300	0.5142
2.1068	2.0	188	2.0110	0.6134	0.6591	0.6134	0.5955
1.8269	3.0	282	1.8153	0.6662	0.6948	0.6662	0.6526
1.6924	4.0	376	1.7128	0.6843	0.7039	0.6843	0.6717
1.4718	5.0	470	1.6318	0.7057	0.7206	0.7057	0.6962
1.4567	6.0	564	1.5817	0.7135	0.7257	0.7135	0.7043
1.3464	7.0	658	1.5305	0.7251	0.7340	0.7251	0.7175
1.265	8.0	752	1.5036	0.7280	0.7358	0.7280	0.7211
1.1838	9.0	846	1.4770	0.7321	0.7405	0.7321	0.7256
1.1135	10.0	940	1.4468	0.7363	0.7424	0.7363	0.7299
1.0988	11.0	1034	1.4392	0.7390	0.7441	0.7390	0.7324
1.0884	12.0	1128	1.4233	0.7406	0.7462	0.7406	0.7348
1.0207	13.0	1222	1.3995	0.7402	0.7450	0.7402	0.7341
1.0033	14.0	1316	1.3999	0.7415	0.7468	0.7415	0.7356
0.965	15.0	1410	1.3832	0.7473	0.7506	0.7473	0.7416
0.9772	16.0	1504	1.3783	0.7459	0.7494	0.7459	0.7401
0.9073	17.0	1598	1.3883	0.7451	0.7504	0.7451	0.7397
0.9094	18.0	1692	1.3710	0.7475	0.7511	0.7475	0.7421
0.9111	19.0	1786	1.3715	0.7472	0.7500	0.7472	0.7415
0.929	20.0	1880	1.3657	0.7470	0.7509	0.7470	0.7419

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.2
• Tokenizers 0.19.1