red-squirrel-detector

This model is a fine-tuned version of ustc-community/dfine-small-coco on the davanstrien/squirrel-cam-labeled dataset. It achieves the following results on the evaluation set:

• Loss: 0.7587
• Map: 0.8614
• Map 50: 0.9087
• Map 75: 0.8839
• Map Small: 0.0
• Map Medium: 0.63
• Map Large: 0.8901
• Mar 1: 0.7528
• Mar 10: 0.8876
• Mar 100: 0.9441
• Mar Small: 0.0
• Mar Medium: 0.7769
• Mar Large: 0.9653
• Map Class 0: 0.8614
• Mar 100 Class 0: 0.9441

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
• num_epochs: 30.0

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 Class 0	Mar 100 Class 0
No log	1.0	91	2.3672	0.1766	0.23	0.1862	0.0	0.0031	0.1989	0.3422	0.5248	0.6224	0.0	0.3462	0.651	0.1766	0.6224
No log	2.0	182	1.0651	0.7244	0.8127	0.7856	0.0	0.4261	0.7538	0.6901	0.7888	0.8596	0.0	0.6385	0.885	0.7244	0.8596
No log	3.0	273	0.8899	0.7577	0.8321	0.7881	0.0	0.4732	0.7986	0.7137	0.8311	0.8863	0.0	0.6769	0.9109	0.7577	0.8863
No log	4.0	364	0.8310	0.7773	0.8432	0.8045	0.0	0.586	0.814	0.7124	0.841	0.8894	0.0	0.7	0.9122	0.7773	0.8894
No log	5.0	455	0.8333	0.7792	0.8513	0.8058	0.0	0.6129	0.8197	0.718	0.8317	0.9031	0.0	0.7077	0.9265	0.7792	0.9031
27.3840	6.0	546	0.8361	0.8008	0.8684	0.8296	0.0	0.5788	0.8348	0.723	0.8609	0.9124	0.0	0.7077	0.9367	0.8008	0.9124
27.3840	7.0	637	0.8078	0.8	0.8587	0.8191	0.0	0.5968	0.8365	0.7224	0.8739	0.9211	0.0	0.7308	0.9442	0.8	0.9211
27.3840	8.0	728	0.8232	0.8173	0.8877	0.837	0.0	0.6199	0.8476	0.7205	0.8677	0.9161	0.0	0.7	0.9415	0.8173	0.9161
27.3840	9.0	819	0.7746	0.8268	0.8863	0.8541	0.0	0.6029	0.8577	0.7211	0.8702	0.9224	0.0	0.6692	0.951	0.8268	0.9224
27.3840	10.0	910	0.8223	0.8324	0.8875	0.8682	0.0	0.6198	0.8586	0.7311	0.8801	0.9193	0.0	0.6923	0.9456	0.8324	0.9193
9.0922	11.0	1001	0.7678	0.8327	0.8851	0.8689	0.0	0.5991	0.8634	0.7422	0.8671	0.9255	0.0	0.6846	0.9531	0.8327	0.9255
9.0922	12.0	1092	0.7850	0.8328	0.8888	0.8545	0.0	0.6134	0.8635	0.7366	0.8752	0.9335	0.0	0.7231	0.9585	0.8328	0.9335
9.0922	13.0	1183	0.8330	0.8294	0.8882	0.8485	0.0	0.583	0.8585	0.7416	0.8689	0.9267	0.0	0.7308	0.9503	0.8294	0.9267
9.0922	14.0	1274	0.7800	0.8399	0.8977	0.8761	0.0056	0.5989	0.87	0.741	0.8758	0.9348	0.6	0.7692	0.9517	0.8399	0.9348
9.0922	15.0	1365	0.7595	0.8437	0.8993	0.8771	0.0061	0.6092	0.8731	0.7509	0.8925	0.9466	0.6	0.7846	0.9633	0.8437	0.9466
9.0922	16.0	1456	0.7758	0.843	0.9059	0.8657	0.0	0.6186	0.8713	0.7472	0.8783	0.9385	0.0	0.7308	0.9633	0.843	0.9385
8.3234	17.0	1547	0.8078	0.844	0.9017	0.8816	0.009	0.6091	0.8754	0.7453	0.8957	0.9366	0.6	0.7077	0.9592	0.844	0.9366
8.3234	18.0	1638	0.7719	0.8479	0.9069	0.8728	0.0	0.6086	0.8785	0.7522	0.8882	0.9273	0.0	0.7077	0.9531	0.8479	0.9273
8.3234	19.0	1729	0.7562	0.8552	0.9105	0.8779	0.0	0.6193	0.8853	0.7509	0.8882	0.9323	0.0	0.7077	0.9585	0.8552	0.9323
8.3234	20.0	1820	0.8166	0.8478	0.9076	0.8774	0.0	0.6428	0.8765	0.7528	0.8988	0.9379	0.0	0.7231	0.9633	0.8478	0.9379
8.3234	21.0	1911	0.7616	0.8519	0.9053	0.8843	0.0	0.6332	0.8818	0.7578	0.8919	0.9385	0.0	0.7154	0.9646	0.8519	0.9385
7.9949	22.0	2002	0.7445	0.857	0.9102	0.8872	0.0	0.6327	0.8856	0.7497	0.8839	0.9354	0.0	0.7231	0.9605	0.857	0.9354
7.9949	23.0	2093	0.7569	0.85	0.9003	0.8792	0.0	0.6308	0.8802	0.7516	0.9019	0.9366	0.0	0.7154	0.9626	0.85	0.9366
7.9949	24.0	2184	0.7550	0.862	0.9093	0.8837	0.0	0.6277	0.8917	0.7516	0.8882	0.9441	0.0	0.7769	0.9653	0.862	0.9441
7.9949	25.0	2275	0.7435	0.8561	0.9053	0.8875	0.0055	0.6371	0.885	0.7559	0.8981	0.9478	0.7	0.7923	0.9633	0.8561	0.9478
7.9949	26.0	2366	0.7752	0.8441	0.8961	0.8669	0.0	0.6392	0.8717	0.7553	0.8857	0.9398	0.0	0.7923	0.9592	0.8441	0.9398
7.9949	27.0	2457	0.7505	0.8505	0.8966	0.8745	0.0	0.6337	0.88	0.7565	0.8857	0.9441	0.0	0.7923	0.9639	0.8505	0.9441
7.6820	28.0	2548	0.7449	0.8533	0.9012	0.8801	0.0	0.627	0.8826	0.7553	0.8901	0.9416	0.0	0.7769	0.9626	0.8533	0.9416
7.6820	29.0	2639	0.7330	0.8577	0.9052	0.8842	0.0076	0.6278	0.8874	0.7571	0.8975	0.9497	0.8	0.7769	0.966	0.8577	0.9497
7.6820	30.0	2730	0.7412	0.8559	0.9016	0.8852	0.0	0.6332	0.8848	0.7571	0.8957	0.9385	0.0	0.7231	0.9639	0.8559	0.9385

Framework versions

• Transformers 5.4.0
• Pytorch 2.11.0+cu130
• Datasets 4.8.4
• Tokenizers 0.22.2