Modelpublic

koppolusameer/rfdetr-medium-basketball-player-referee-detection

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

0likes20downloads

Model Card

rfdetr-medium-basketball-player-referee-detection

This model is a fine-tuned version of Roboflow/rf-detr-medium on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 16.7540
—Map: 0.7297
—Map 50: 0.971
—Map 75: 0.8533
—Map Small: 0.5023
—Map Medium: 0.7305
—Map Large: 0.8004
—Mar 1: 0.1932
—Mar 10: 0.7558
—Mar 100: 0.7909
—Mar Small: 0.8
—Mar Medium: 0.7907
—Mar Large: 0.8111
—Map Player: 0.7316
—Map Referee: 0.7277

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: 2e-05
—trainbatchsize: 2
—evalbatchsize: 2
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: cosine
—num_epochs: 20

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 Player	Map Referee
7.3711	1.0	232	14.3762	0.4327	0.6763	0.4951	0.5	0.4338	0.4	0.1226	0.6153	0.7532	0.5	0.7545	0.5333	0.559	0.3064
6.0851	2.0	464	15.3272	0.6024	0.934	0.7044	0.5	0.6035	0.5358	0.1629	0.6584	0.7392	0.5	0.7396	0.7222	0.6316	0.5731
5.1302	3.0	696	16.1523	0.6558	0.9628	0.7659	0.6038	0.6568	0.7101	0.172	0.6964	0.7716	0.7	0.7717	0.7778	0.663	0.6487
4.7785	4.0	928	15.6558	0.6815	0.9556	0.8144	0.5	0.6833	0.6063	0.1811	0.7109	0.7817	0.5	0.7821	0.7222	0.6864	0.6766
4.8138	5.0	1160	16.0388	0.6704	0.9485	0.7884	0.7034	0.6715	0.5819	0.1812	0.701	0.7606	0.9	0.7608	0.7111	0.684	0.6568
4.2065	6.0	1392	16.0623	0.656	0.9382	0.7873	0.1034	0.6572	0.706	0.1807	0.6862	0.7469	0.4	0.7469	0.8	0.6734	0.6386
4.1001	7.0	1624	16.2022	0.6798	0.9254	0.8015	0.5	0.6809	0.582	0.1885	0.7009	0.7632	0.5	0.7635	0.7556	0.7091	0.6504
3.7894	8.0	1856	16.3419	0.6659	0.9089	0.7841	0.503	0.6666	0.7698	0.1894	0.6876	0.7452	0.7	0.7451	0.8111	0.6944	0.6373
3.7885	9.0	2088	15.8034	0.6903	0.9424	0.8142	0.5029	0.6914	0.6873	0.1883	0.7166	0.7728	0.8	0.7728	0.7778	0.7028	0.6777
3.7385	10.0	2320	16.1087	0.7012	0.9634	0.8318	0.6	0.7021	0.6731	0.1855	0.731	0.7769	0.6	0.7772	0.7333	0.716	0.6865
3.5390	11.0	2552	16.6688	0.71	0.9635	0.831	0.4033	0.7113	0.691	0.1926	0.7385	0.7852	0.6	0.7852	0.8	0.7154	0.7045
3.4166	12.0	2784	16.1298	0.7232	0.9746	0.8545	0.601	0.7238	0.7877	0.1914	0.7526	0.787	0.7	0.787	0.8	0.7231	0.7233
2.9186	13.0	3016	16.4147	0.7121	0.964	0.8429	0.5022	0.713	0.7383	0.1904	0.7402	0.7782	0.7	0.7783	0.7889	0.7215	0.7027
3.3311	14.0	3248	16.6527	0.7206	0.9657	0.8435	0.6	0.7213	0.8017	0.1914	0.7473	0.7813	0.6	0.7812	0.8222	0.7306	0.7106
3.2214	15.0	3480	16.7827	0.7243	0.9657	0.8497	0.5005	0.7252	0.7971	0.1938	0.7527	0.7891	0.6	0.7892	0.8	0.7341	0.7146
3.1278	16.0	3712	16.7533	0.7325	0.9726	0.8536	0.5015	0.7333	0.7879	0.1948	0.7574	0.792	0.7	0.7921	0.7889	0.734	0.7311
2.9767	17.0	3944	16.8603	0.7211	0.9685	0.847	0.4016	0.7218	0.7772	0.1917	0.75	0.7829	0.6	0.783	0.7889	0.7261	0.7161
2.7972	18.0	4176	16.7250	0.7274	0.9705	0.855	0.6	0.7281	0.7801	0.1923	0.7538	0.7904	0.6	0.7905	0.8	0.7315	0.7232
2.8994	19.0	4408	16.7552	0.7276	0.9685	0.8523	0.6008	0.7284	0.8012	0.1928	0.7551	0.7911	0.7	0.791	0.8111	0.7315	0.7237
2.9083	20.0	4640	16.7638	0.7296	0.971	0.8524	0.5023	0.7304	0.8004	0.193	0.7555	0.7905	0.8	0.7904	0.8111	0.7313	0.728

Framework versions

—Transformers 5.13.1
—Pytorch 2.11.0+cu128
—Datasets 4.0.0
—Tokenizers 0.22.2