This model is a fine-tuned version of microsoft/conditional-detr-resnet-50 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.0912
Map: 0.9334
Map 50: 0.9684
Map 75: 0.9684
Map Small: -1.0
Map Medium: 0.9334
Map Large: -1.0
Mar 1: 0.0125
Mar 10: 0.1259
Mar 100: 0.9777
Mar Small: -1.0
Mar Medium: 0.9777
Mar Large: -1.0
Map Uled: 0.9334
Mar 100 Uled: 0.9777
Map Trash: -1.0
Mar 100 Trash: -1.0

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: 32
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: 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 Uled	Mar 100 Uled	Map Trash	Mar 100 Trash
No log	1.0	41	0.2460	0.7925	0.9619	0.9382	-1.0	0.7925	-1.0	0.0115	0.1133	0.8652	-1.0	0.8652	-1.0	0.7925	0.8652	-1.0	-1.0
No log	2.0	82	0.2123	0.8121	0.9671	0.9527	-1.0	0.8121	-1.0	0.0111	0.1125	0.8797	-1.0	0.8797	-1.0	0.8121	0.8797	-1.0	-1.0
No log	3.0	123	0.1597	0.8576	0.9645	0.963	-1.0	0.8576	-1.0	0.0118	0.1181	0.9217	-1.0	0.9217	-1.0	0.8576	0.9217	-1.0	-1.0
No log	4.0	164	0.1645	0.8532	0.9644	0.9606	-1.0	0.8532	-1.0	0.0118	0.1184	0.9174	-1.0	0.9174	-1.0	0.8532	0.9174	-1.0	-1.0
No log	5.0	205	0.2037	0.824	0.9632	0.9614	-1.0	0.824	-1.0	0.0115	0.1142	0.8826	-1.0	0.8826	-1.0	0.824	0.8826	-1.0	-1.0
No log	6.0	246	0.1342	0.8864	0.9672	0.9665	-1.0	0.8864	-1.0	0.0119	0.1213	0.9429	-1.0	0.9429	-1.0	0.8864	0.9429	-1.0	-1.0
No log	7.0	287	0.1365	0.8821	0.9677	0.9672	-1.0	0.8821	-1.0	0.0121	0.1218	0.9362	-1.0	0.9362	-1.0	0.8821	0.9362	-1.0	-1.0
No log	8.0	328	0.1470	0.872	0.9666	0.9662	-1.0	0.872	-1.0	0.0119	0.12	0.9326	-1.0	0.9326	-1.0	0.872	0.9326	-1.0	-1.0
No log	9.0	369	0.1783	0.8495	0.9678	0.9673	-1.0	0.8495	-1.0	0.0118	0.118	0.9017	-1.0	0.9017	-1.0	0.8495	0.9017	-1.0	-1.0
No log	10.0	410	0.1563	0.8676	0.9662	0.9643	-1.0	0.8676	-1.0	0.012	0.1203	0.9225	-1.0	0.9225	-1.0	0.8676	0.9225	-1.0	-1.0
No log	11.0	451	0.1458	0.8783	0.966	0.9658	-1.0	0.8783	-1.0	0.012	0.121	0.9321	-1.0	0.9321	-1.0	0.8783	0.9321	-1.0	-1.0
No log	12.0	492	0.1273	0.8939	0.9669	0.9667	-1.0	0.8939	-1.0	0.0123	0.1234	0.9462	-1.0	0.9462	-1.0	0.8939	0.9462	-1.0	-1.0
0.2348	13.0	533	0.1376	0.8862	0.9683	0.968	-1.0	0.8862	-1.0	0.0121	0.1217	0.9404	-1.0	0.9404	-1.0	0.8862	0.9404	-1.0	-1.0
0.2348	14.0	574	0.1338	0.8865	0.9669	0.9668	-1.0	0.8865	-1.0	0.0122	0.1222	0.9422	-1.0	0.9422	-1.0	0.8865	0.9422	-1.0	-1.0
0.2348	15.0	615	0.1258	0.8917	0.9685	0.9685	-1.0	0.8917	-1.0	0.012	0.1221	0.9454	-1.0	0.9454	-1.0	0.8917	0.9454	-1.0	-1.0
0.2348	16.0	656	0.1206	0.8998	0.9689	0.9689	-1.0	0.8998	-1.0	0.0123	0.1233	0.9524	-1.0	0.9524	-1.0	0.8998	0.9524	-1.0	-1.0
0.2348	17.0	697	0.1075	0.911	0.969	0.969	-1.0	0.911	-1.0	0.0123	0.1238	0.9612	-1.0	0.9612	-1.0	0.911	0.9612	-1.0	-1.0
0.2348	18.0	738	0.1084	0.9113	0.9692	0.9691	-1.0	0.9113	-1.0	0.0123	0.1237	0.9628	-1.0	0.9628	-1.0	0.9113	0.9628	-1.0	-1.0
0.2348	19.0	779	0.1104	0.91	0.9688	0.9688	-1.0	0.91	-1.0	0.0123	0.1236	0.9602	-1.0	0.9602	-1.0	0.91	0.9602	-1.0	-1.0
0.2348	20.0	820	0.1097	0.9103	0.9693	0.9693	-1.0	0.9103	-1.0	0.0123	0.1241	0.9616	-1.0	0.9616	-1.0	0.9103	0.9616	-1.0	-1.0
0.2348	21.0	861	0.1111	0.9106	0.9666	0.9665	-1.0	0.9106	-1.0	0.0123	0.1242	0.9624	-1.0	0.9624	-1.0	0.9106	0.9624	-1.0	-1.0
0.2348	22.0	902	0.1007	0.923	0.9667	0.9666	-1.0	0.923	-1.0	0.0125	0.1251	0.972	-1.0	0.972	-1.0	0.923	0.972	-1.0	-1.0
0.2348	23.0	943	0.1080	0.9103	0.9671	0.9671	-1.0	0.9103	-1.0	0.0123	0.1242	0.9612	-1.0	0.9612	-1.0	0.9103	0.9612	-1.0	-1.0
0.2348	24.0	984	0.0987	0.9197	0.967	0.967	-1.0	0.9197	-1.0	0.0124	0.1253	0.9697	-1.0	0.9697	-1.0	0.9197	0.9697	-1.0	-1.0
0.1648	25.0	1025	0.0979	0.9226	0.9675	0.9675	-1.0	0.9226	-1.0	0.0125	0.1253	0.9715	-1.0	0.9715	-1.0	0.9226	0.9715	-1.0	-1.0
0.1648	26.0	1066	0.0912	0.9334	0.9684	0.9684	-1.0	0.9334	-1.0	0.0125	0.1259	0.9777	-1.0	0.9777	-1.0	0.9334	0.9777	-1.0	-1.0
0.1648	27.0	1107	0.0926	0.9311	0.9682	0.9682	-1.0	0.9311	-1.0	0.0125	0.1258	0.9763	-1.0	0.9763	-1.0	0.9311	0.9763	-1.0	-1.0
0.1648	28.0	1148	0.0933	0.9301	0.9682	0.9681	-1.0	0.9301	-1.0	0.0125	0.1258	0.9756	-1.0	0.9756	-1.0	0.9301	0.9756	-1.0	-1.0
0.1648	29.0	1189	0.0937	0.9301	0.9682	0.9681	-1.0	0.9301	-1.0	0.0125	0.1259	0.9758	-1.0	0.9758	-1.0	0.9301	0.9758	-1.0	-1.0
0.1648	30.0	1230	0.0932	0.9311	0.9682	0.9681	-1.0	0.9311	-1.0	0.0125	0.126	0.9763	-1.0	0.9763	-1.0	0.9311	0.9763	-1.0	-1.0

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0

pylu5229/conditional-detr-resnet-50-uLED-obj-detect-test

conditional-detr-resnet-50-uLED-obj-detect-test

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

pylu5229/
conditional-detr-resnet-50-uLED-obj-detect-test