mobilenetv3-BlurryDetection-v4

This model is a fine-tuned version of MinhLe999/mobilenetv3-HandwritingStrip-RandImg on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0436
• Accuracy: 0.9875
• Precision: 0.9957
• Recall: 0.9777
• F1: 0.9866
• Roc Auc: 0.9994

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: 0.0003
• train_batch_size: 32
• eval_batch_size: 32
• 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: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Roc Auc
0.0805	0.4405	200	0.1072	0.9685	1.0	0.9330	0.9653	0.9986
0.0931	0.8811	400	0.0377	0.988	0.9914	0.9830	0.9872	0.9993
0.0750	1.3216	600	0.0352	0.988	0.9841	0.9904	0.9873	0.9992
0.0469	1.7621	800	0.0525	0.985	0.9946	0.9734	0.9839	0.9987
0.0254	2.2026	1000	0.1051	0.976	0.9945	0.9543	0.9739	0.9975
0.0415	2.6432	1200	0.0577	0.9815	0.9967	0.9638	0.9800	0.9990
0.0059	3.0837	1400	0.0591	0.9815	0.9978	0.9628	0.9800	0.9991
0.0106	3.5242	1600	0.0577	0.9805	0.9956	0.9628	0.9789	0.9993
0.0295	3.9648	1800	0.0436	0.9875	0.9957	0.9777	0.9866	0.9994

Framework versions

• Transformers 5.3.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2