BxSD-finetuned-UCF_RGBCROP_5e6-poly

This model is a fine-tuned version of TanAlexanderlz/BxSD_RGBCROP_Aug16F-8B16F5e6-poly on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3295
• Accuracy: 0.8056

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: 8
• eval_batch_size: 8
• seed: 42
• 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: polynomial
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 345

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5386	0.0696	24	0.5426	0.7619
0.2841	1.0696	48	0.4681	0.8333
0.0153	2.0696	72	0.3577	0.9048
0.0034	3.0696	96	0.4862	0.9048
0.0014	4.0696	120	0.4587	0.9048
0.0011	5.0696	144	0.4434	0.9286
0.0009	6.0696	168	0.4669	0.9286
0.0008	7.0696	192	0.4841	0.9286
0.0007	8.0696	216	0.4640	0.9286
0.0006	9.0696	240	0.4720	0.9286
0.0006	10.0696	264	0.4727	0.9286
0.0006	11.0696	288	0.4804	0.9286
0.0005	12.0696	312	0.4818	0.9286
0.0005	13.0696	336	0.4828	0.9286
0.0006	14.0261	345	0.4825	0.9286

Framework versions

• Transformers 4.52.2
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.1