yolos-small-license-plates

This model is a fine-tuned version of hustvl/yolos-small on the merve/license-plates dataset. It achieves the following results on the evaluation set:

Loss: 0.6225
Map: 0.534
Map 50: 0.9077
Map 75: 0.5625
Map Small: 0.2792
Map Medium: 0.6175
Map Large: 0.5909
Mar 1: 0.5969
Mar 10: 0.6588
Mar 100: 0.6656
Mar Small: 0.4416
Mar Medium: 0.739
Mar Large: 0.7841
Map License Plate: 0.534
Mar 100 License Plate: 0.6656

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: 5e-05
train_batch_size: 8
eval_batch_size: 8
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
lr_scheduler_warmup_steps: 0.1
num_epochs: 10.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map License Plate	Mar 100 License Plate
0.9665	1.0	78	0.8866	0.1357	0.2747	0.1165	0.0512	0.1673	0.2354	0.4036	0.5571	0.5781	0.3163	0.6675	0.6591	0.1357	0.5781
0.8946	2.0	156	0.9421	0.127	0.3018	0.0788	0.0531	0.1545	0.1969	0.335	0.5091	0.5127	0.2767	0.5907	0.6273	0.127	0.5127
0.9292	3.0	234	0.8366	0.1821	0.3682	0.1563	0.0482	0.2295	0.195	0.4215	0.5705	0.5764	0.2882	0.6706	0.7318	0.1821	0.5764
0.8067	4.0	312	0.8124	0.2511	0.5173	0.2042	0.0702	0.3177	0.2809	0.445	0.5582	0.5638	0.2722	0.6586	0.7273	0.2511	0.5638
0.8869	5.0	390	0.8012	0.3589	0.7432	0.2985	0.1297	0.4457	0.3999	0.4716	0.5969	0.6014	0.3335	0.6876	0.7682	0.3589	0.6014
0.7027	6.0	468	0.7083	0.4386	0.8228	0.4143	0.1773	0.5458	0.5177	0.5388	0.6156	0.6212	0.3816	0.6976	0.7795	0.4386	0.6212
0.6596	7.0	546	0.7296	0.4446	0.8609	0.3926	0.1807	0.5398	0.4826	0.523	0.6069	0.6114	0.3714	0.6905	0.7295	0.4446	0.6114
0.5702	8.0	624	0.6542	0.5124	0.8907	0.5402	0.2531	0.6015	0.5821	0.5764	0.645	0.6504	0.3992	0.7341	0.7614	0.5124	0.6504
0.6034	9.0	702	0.6305	0.5285	0.9031	0.5595	0.2685	0.6143	0.58	0.591	0.654	0.661	0.4388	0.7326	0.7955	0.5285	0.661
0.5219	10.0	780	0.6225	0.534	0.9077	0.5625	0.2792	0.6175	0.5909	0.5969	0.6588	0.6656	0.4416	0.739	0.7841	0.534	0.6656

Framework versions

Transformers 5.3.0.dev0
Pytorch 2.10.0+cu128
Datasets 4.6.0
Tokenizers 0.22.2

Loading…

yolos-small-license-plates

This model is a fine-tuned version of hustvl/yolos-small on the merve/license-plates dataset. It achieves the following results on the evaluation set:

Loss: 0.6225

Map: 0.534

Map 50: 0.9077

Map 75: 0.5625

Map Small: 0.2792

Map Medium: 0.6175

Map Large: 0.5909

Mar 1: 0.5969

Mar 10: 0.6588

Mar 100: 0.6656

Mar Small: 0.4416

Mar Medium: 0.739

Mar Large: 0.7841

Map License Plate: 0.534

Mar 100 License Plate: 0.6656

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05

train_batch_size: 8

eval_batch_size: 8

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

lr_scheduler_warmup_steps: 0.1

num_epochs: 10.0

mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map License Plate	Mar 100 License Plate
0.9665	1.0	78	0.8866	0.1357	0.2747	0.1165	0.0512	0.1673	0.2354	0.4036	0.5571	0.5781	0.3163	0.6675	0.6591	0.1357	0.5781
0.8946	2.0	156	0.9421	0.127	0.3018	0.0788	0.0531	0.1545	0.1969	0.335	0.5091	0.5127	0.2767	0.5907	0.6273	0.127	0.5127
0.9292	3.0	234	0.8366	0.1821	0.3682	0.1563	0.0482	0.2295	0.195	0.4215	0.5705	0.5764	0.2882	0.6706	0.7318	0.1821	0.5764
0.8067	4.0	312	0.8124	0.2511	0.5173	0.2042	0.0702	0.3177	0.2809	0.445	0.5582	0.5638	0.2722	0.6586	0.7273	0.2511	0.5638
0.8869	5.0	390	0.8012	0.3589	0.7432	0.2985	0.1297	0.4457	0.3999	0.4716	0.5969	0.6014	0.3335	0.6876	0.7682	0.3589	0.6014
0.7027	6.0	468	0.7083	0.4386	0.8228	0.4143	0.1773	0.5458	0.5177	0.5388	0.6156	0.6212	0.3816	0.6976	0.7795	0.4386	0.6212
0.6596	7.0	546	0.7296	0.4446	0.8609	0.3926	0.1807	0.5398	0.4826	0.523	0.6069	0.6114	0.3714	0.6905	0.7295	0.4446	0.6114
0.5702	8.0	624	0.6542	0.5124	0.8907	0.5402	0.2531	0.6015	0.5821	0.5764	0.645	0.6504	0.3992	0.7341	0.7614	0.5124	0.6504
0.6034	9.0	702	0.6305	0.5285	0.9031	0.5595	0.2685	0.6143	0.58	0.591	0.654	0.661	0.4388	0.7326	0.7955	0.5285	0.661
0.5219	10.0	780	0.6225	0.534	0.9077	0.5625	0.2792	0.6175	0.5909	0.5969	0.6588	0.6656	0.4416	0.739	0.7841	0.534	0.6656

Framework versions

Transformers 5.3.0.dev0

Pytorch 2.10.0+cu128

Datasets 4.6.0

Tokenizers 0.22.2

merve/yolos-small-license-plates

yolos-small-license-plates

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

yolos-small-license-plates

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions