klasifikasiburung_new

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: 1.1769
• Accuracy: 0.7604
• Precision: 0.7654
• Recall: 0.7604
• F1: 0.7572

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: 16
• eval_batch_size: 16
• 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
2.3725	1.0	375	2.1701	0.5720	0.6410	0.5720	0.5531
1.9971	2.0	750	1.7855	0.6595	0.6896	0.6595	0.6456
1.6092	3.0	1125	1.5948	0.7026	0.7201	0.7026	0.6921
1.5044	4.0	1500	1.4862	0.7173	0.7288	0.7173	0.7078
1.2893	5.0	1875	1.4145	0.7309	0.7402	0.7309	0.7236
1.2276	6.0	2250	1.3653	0.7373	0.7454	0.7373	0.7310
1.1467	7.0	2625	1.3099	0.7478	0.7536	0.7478	0.7420
1.0491	8.0	3000	1.2975	0.7451	0.7518	0.7451	0.7399
0.9231	9.0	3375	1.2683	0.7518	0.7574	0.7518	0.7470
0.8979	10.0	3750	1.2389	0.7561	0.7609	0.7561	0.7519
0.9467	11.0	4125	1.2400	0.7566	0.7608	0.7566	0.7517
0.8315	12.0	4500	1.2164	0.7565	0.7623	0.7565	0.7530
0.7316	13.0	4875	1.2005	0.7570	0.7612	0.7570	0.7531
0.6786	14.0	5250	1.2080	0.7560	0.7623	0.7560	0.7527
0.7923	15.0	5625	1.1869	0.7582	0.7628	0.7582	0.7545
0.7415	16.0	6000	1.1802	0.7575	0.7633	0.7575	0.7548
0.6292	17.0	6375	1.1994	0.7542	0.7602	0.7542	0.7513
0.7069	18.0	6750	1.1769	0.7604	0.7654	0.7604	0.7572
0.69	19.0	7125	1.1743	0.7572	0.7610	0.7572	0.7542
0.6476	20.0	7500	1.1704	0.7585	0.7638	0.7585	0.7558

Framework versions

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