yolo_finetuned_kangaroo

This model is a fine-tuned version of hustvl/yolos-tiny on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8055
• Map: 0.5951
• Map 50: 0.903
• Map 75: 0.6687
• Map Small: -1.0
• Map Medium: 0.3896
• Map Large: 0.6104
• Mar 1: 0.4196
• Mar 10: 0.7125
• Mar 100: 0.7625
• Mar Small: -1.0
• Mar Medium: 0.6
• Mar Large: 0.7717
• Map Kangaroo: 0.5951
• Mar 100 Kangaroo: 0.7625

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: 4
• 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: cosine
• num_epochs: 30

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 Kangaroo	Mar 100 Kangaroo
No log	1.0	33	0.9143	0.4463	0.7466	0.4805	-1.0	0.2209	0.4608	0.3536	0.6446	0.7446	-1.0	0.5333	0.7566	0.4463	0.7446
No log	2.0	66	0.9281	0.4662	0.7555	0.5234	-1.0	0.2095	0.4827	0.3464	0.65	0.7179	-1.0	0.5	0.7302	0.4662	0.7179
No log	3.0	99	0.9294	0.4739	0.8018	0.5279	-1.0	0.3324	0.4864	0.3411	0.6643	0.7286	-1.0	0.5333	0.7396	0.4739	0.7286
No log	4.0	132	0.9755	0.4551	0.7987	0.4899	-1.0	0.2318	0.4696	0.3429	0.6339	0.6893	-1.0	0.5	0.7	0.4551	0.6893
No log	5.0	165	0.9167	0.4733	0.8054	0.5324	-1.0	0.2674	0.4887	0.3554	0.6518	0.7143	-1.0	0.5667	0.7226	0.4733	0.7143
No log	6.0	198	0.8466	0.5278	0.852	0.5946	-1.0	0.4534	0.5381	0.3554	0.6804	0.7411	-1.0	0.6667	0.7453	0.5278	0.7411
No log	7.0	231	0.8655	0.511	0.8133	0.5681	-1.0	0.4087	0.5217	0.3661	0.6679	0.7286	-1.0	0.7333	0.7283	0.511	0.7286
No log	8.0	264	0.8606	0.5317	0.8302	0.592	-1.0	0.2596	0.5459	0.3804	0.6696	0.725	-1.0	0.5	0.7377	0.5317	0.725
No log	9.0	297	0.9685	0.4836	0.8608	0.486	-1.0	0.327	0.4949	0.3696	0.6446	0.6857	-1.0	0.6667	0.6868	0.4836	0.6857
No log	10.0	330	0.8259	0.5465	0.8469	0.6477	-1.0	0.2736	0.5618	0.3821	0.6929	0.7375	-1.0	0.6667	0.7415	0.5465	0.7375
No log	11.0	363	0.8854	0.5498	0.8708	0.5409	-1.0	0.3311	0.5637	0.3911	0.6929	0.725	-1.0	0.6	0.7321	0.5498	0.725
No log	12.0	396	0.8918	0.5225	0.8388	0.5895	-1.0	0.1666	0.5553	0.3786	0.6679	0.7214	-1.0	0.4	0.7396	0.5225	0.7214
No log	13.0	429	0.8681	0.5524	0.8552	0.6257	-1.0	0.2682	0.5655	0.4125	0.6946	0.75	-1.0	0.5667	0.7604	0.5524	0.75
No log	14.0	462	0.8663	0.574	0.8733	0.6568	-1.0	0.2556	0.5915	0.4054	0.7	0.7482	-1.0	0.6667	0.7528	0.574	0.7482
No log	15.0	495	0.8300	0.5832	0.885	0.6767	-1.0	0.4304	0.5929	0.4196	0.7107	0.7625	-1.0	0.6667	0.7679	0.5832	0.7625
0.5719	16.0	528	0.8050	0.5825	0.8845	0.6668	-1.0	0.3807	0.5972	0.4071	0.7107	0.7696	-1.0	0.6	0.7792	0.5825	0.7696
0.5719	17.0	561	0.8186	0.5979	0.9085	0.697	-1.0	0.4183	0.6086	0.4268	0.7143	0.7571	-1.0	0.6667	0.7623	0.5979	0.7571
0.5719	18.0	594	0.8352	0.5901	0.9151	0.66	-1.0	0.336	0.6044	0.4196	0.7071	0.7464	-1.0	0.6	0.7547	0.5901	0.7464
0.5719	19.0	627	0.8484	0.5815	0.8982	0.6738	-1.0	0.398	0.5932	0.4071	0.7054	0.7571	-1.0	0.6333	0.7642	0.5815	0.7571
0.5719	20.0	660	0.8100	0.5853	0.896	0.6717	-1.0	0.2612	0.6055	0.4196	0.7125	0.7607	-1.0	0.5667	0.7717	0.5853	0.7607
0.5719	21.0	693	0.7945	0.5982	0.9116	0.7068	-1.0	0.3321	0.616	0.4179	0.7179	0.7589	-1.0	0.6	0.7679	0.5982	0.7589
0.5719	22.0	726	0.8120	0.5874	0.8983	0.6865	-1.0	0.3651	0.6001	0.4089	0.7161	0.7571	-1.0	0.6333	0.7642	0.5874	0.7571
0.5719	23.0	759	0.8185	0.5949	0.9052	0.6507	-1.0	0.3641	0.6102	0.4196	0.7125	0.7536	-1.0	0.6	0.7623	0.5949	0.7536
0.5719	24.0	792	0.8210	0.5925	0.9068	0.666	-1.0	0.3782	0.6086	0.4125	0.7089	0.7625	-1.0	0.6333	0.7698	0.5925	0.7625
0.5719	25.0	825	0.8041	0.5894	0.8991	0.6653	-1.0	0.3798	0.6049	0.4161	0.7179	0.7625	-1.0	0.6	0.7717	0.5894	0.7625
0.5719	26.0	858	0.8112	0.5939	0.9058	0.6762	-1.0	0.3884	0.6087	0.4179	0.7161	0.7589	-1.0	0.6333	0.766	0.5939	0.7589
0.5719	27.0	891	0.8068	0.5976	0.9036	0.6688	-1.0	0.4224	0.6119	0.4179	0.7143	0.7625	-1.0	0.6	0.7717	0.5976	0.7625
0.5719	28.0	924	0.8057	0.5948	0.9037	0.6668	-1.0	0.3896	0.6102	0.4196	0.7125	0.7607	-1.0	0.6	0.7698	0.5948	0.7607
0.5719	29.0	957	0.8055	0.5951	0.903	0.6687	-1.0	0.3896	0.6104	0.4196	0.7125	0.7625	-1.0	0.6	0.7717	0.5951	0.7625
0.5719	30.0	990	0.8055	0.5951	0.903	0.6687	-1.0	0.3896	0.6104	0.4196	0.7125	0.7625	-1.0	0.6	0.7717	0.5951	0.7625

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2