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 Raccoon	Mar 100 Raccoon
No log	1.0	40	0.8582	0.5533	0.8928	0.5718	-1.0	0.2302	0.6053	0.569	0.7548	0.8286	-1.0	0.72	0.8432	0.5533	0.8286
No log	2.0	80	0.8716	0.5902	0.9128	0.6887	-1.0	0.3087	0.63	0.5976	0.7357	0.8024	-1.0	0.58	0.8324	0.5902	0.8024
No log	3.0	120	0.9247	0.5676	0.8867	0.6475	-1.0	0.249	0.613	0.5429	0.7286	0.7905	-1.0	0.62	0.8135	0.5676	0.7905
No log	4.0	160	0.9760	0.5413	0.8737	0.6177	-1.0	0.211	0.6	0.5548	0.6881	0.7929	-1.0	0.52	0.8297	0.5413	0.7929
No log	5.0	200	0.7318	0.6146	0.9072	0.7043	-1.0	0.2886	0.6737	0.6429	0.7714	0.831	-1.0	0.64	0.8568	0.6146	0.831
No log	6.0	240	0.7990	0.567	0.8684	0.6281	-1.0	0.2244	0.6424	0.5833	0.7381	0.7952	-1.0	0.66	0.8135	0.567	0.7952
No log	7.0	280	0.7510	0.6007	0.8624	0.6888	-1.0	0.3138	0.6528	0.6119	0.7667	0.8286	-1.0	0.72	0.8432	0.6007	0.8286
No log	8.0	320	0.7264	0.6517	0.9172	0.8309	-1.0	0.379	0.6937	0.65	0.7905	0.831	-1.0	0.74	0.8432	0.6517	0.831
No log	9.0	360	0.7492	0.614	0.8863	0.7064	-1.0	0.4346	0.6448	0.6214	0.7976	0.8262	-1.0	0.74	0.8378	0.614	0.8262
No log	10.0	400	0.7869	0.6067	0.891	0.6556	-1.0	0.3134	0.6606	0.5952	0.7786	0.819	-1.0	0.66	0.8405	0.6067	0.819
No log	11.0	440	0.7511	0.6029	0.8962	0.6608	-1.0	0.3902	0.6436	0.6619	0.7833	0.8286	-1.0	0.74	0.8405	0.6029	0.8286
No log	12.0	480	0.7275	0.6454	0.9143	0.7474	-1.0	0.3872	0.692	0.6381	0.8095	0.8357	-1.0	0.64	0.8622	0.6454	0.8357
0.5573	13.0	520	0.7762	0.6216	0.9067	0.7235	-1.0	0.3677	0.6627	0.6262	0.8048	0.8214	-1.0	0.66	0.8432	0.6216	0.8214
0.5573	14.0	560	0.7516	0.6261	0.9312	0.6971	-1.0	0.4364	0.6561	0.6595	0.7929	0.8357	-1.0	0.74	0.8486	0.6261	0.8357
0.5573	15.0	600	0.7513	0.6174	0.9096	0.7606	-1.0	0.3891	0.6531	0.6119	0.8048	0.831	-1.0	0.7	0.8486	0.6174	0.831
0.5573	16.0	640	0.7384	0.6226	0.8896	0.7249	-1.0	0.397	0.6585	0.6119	0.8024	0.8429	-1.0	0.7	0.8622	0.6226	0.8429
0.5573	17.0	680	0.7254	0.6185	0.8767	0.7185	-1.0	0.3926	0.6511	0.6167	0.7976	0.8357	-1.0	0.66	0.8595	0.6185	0.8357
0.5573	18.0	720	0.7359	0.6325	0.8808	0.7802	-1.0	0.4496	0.6607	0.6238	0.8214	0.8571	-1.0	0.8	0.8649	0.6325	0.8571
0.5573	19.0	760	0.7169	0.6272	0.8713	0.6995	-1.0	0.3902	0.6687	0.6595	0.8024	0.8429	-1.0	0.76	0.8541	0.6272	0.8429
0.5573	20.0	800	0.7226	0.6417	0.9118	0.768	-1.0	0.4189	0.6786	0.6262	0.8119	0.8452	-1.0	0.74	0.8595	0.6417	0.8452
0.5573	21.0	840	0.7260	0.6448	0.9079	0.7485	-1.0	0.4408	0.6764	0.6238	0.8333	0.8548	-1.0	0.78	0.8649	0.6448	0.8548
0.5573	22.0	880	0.7427	0.637	0.9148	0.7791	-1.0	0.3805	0.677	0.6167	0.8	0.8429	-1.0	0.74	0.8568	0.637	0.8429
0.5573	23.0	920	0.7308	0.6338	0.8947	0.7758	-1.0	0.3629	0.678	0.6214	0.8095	0.8429	-1.0	0.76	0.8541	0.6338	0.8429
0.5573	24.0	960	0.7138	0.6464	0.8921	0.807	-1.0	0.4115	0.6814	0.6143	0.819	0.8405	-1.0	0.72	0.8568	0.6464	0.8405
0.3623	25.0	1000	0.7254	0.6413	0.8959	0.805	-1.0	0.4044	0.6769	0.6	0.8095	0.8357	-1.0	0.7	0.8541	0.6413	0.8357
0.3623	26.0	1040	0.7147	0.636	0.8825	0.7854	-1.0	0.3913	0.6708	0.6	0.8048	0.8357	-1.0	0.72	0.8514	0.636	0.8357
0.3623	27.0	1080	0.7169	0.6387	0.8887	0.7882	-1.0	0.4061	0.6729	0.5952	0.7976	0.831	-1.0	0.72	0.8459	0.6387	0.831
0.3623	28.0	1120	0.7134	0.6404	0.886	0.7963	-1.0	0.4204	0.6729	0.6024	0.8024	0.831	-1.0	0.72	0.8459	0.6404	0.831
0.3623	29.0	1160	0.7131	0.6411	0.8875	0.7982	-1.0	0.4195	0.6744	0.6024	0.8024	0.831	-1.0	0.72	0.8459	0.6411	0.831
0.3623	30.0	1200	0.7133	0.6414	0.8873	0.7986	-1.0	0.4196	0.6749	0.6024	0.8024	0.831	-1.0	0.72	0.8459	0.6414	0.831

Framework versions

Transformers 4.57.6
Pytorch 2.9.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

alperema/yolo_finetuned_raccoon

yolo_finetuned_raccoon

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

alperema/
yolo_finetuned_raccoon