chickens-60-epoch-200-images

This model is a fine-tuned version of facebook/detr-resnet-50 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5714
• Map: 0.6874
• Map 50: 0.9234
• Map 75: 0.8248
• Map Small: 0.5069
• Map Medium: 0.6714
• Map Large: 0.7654
• Mar 1: 0.2397
• Mar 10: 0.7424
• Mar 100: 0.7555
• Mar Small: 0.5185
• Mar Medium: 0.7332
• Mar Large: 0.8229
• Map Chicken: 0.7156
• Mar 100 Chicken: 0.7929
• Map Duck: 0.6392
• Mar 100 Duck: 0.7115
• Map Plant: 0.7073
• Mar 100 Plant: 0.7621

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: 1e-05
• train_batch_size: 2
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 60

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 Chicken	Mar 100 Chicken	Map Duck	Mar 100 Duck	Map Plant	Mar 100 Plant
1.8277	1.0	100	1.7954	0.0273	0.0429	0.0322	0.0023	0.0189	0.1724	0.0322	0.0974	0.252	0.1333	0.2442	0.3118	0.0101	0.4325	0.0066	0.0358	0.0651	0.2877
1.6529	2.0	200	1.6153	0.0747	0.1142	0.0884	0.0115	0.0385	0.3795	0.0361	0.1619	0.2907	0.1407	0.2683	0.6732	0.0202	0.2592	0.0	0.0	0.2039	0.6129
1.5757	3.0	300	1.5640	0.1063	0.1635	0.12	0.152	0.0606	0.4749	0.0447	0.2034	0.2538	0.2444	0.2323	0.6948	0.0348	0.1246	0.0	0.0	0.2842	0.6369
1.3741	4.0	400	1.4431	0.1124	0.1636	0.1307	0.0213	0.0532	0.5408	0.0402	0.1769	0.2496	0.2259	0.2309	0.7725	0.015	0.0296	0.0	0.0	0.3222	0.7191
1.3922	5.0	500	1.3206	0.1518	0.2268	0.1758	0.1359	0.0896	0.5684	0.051	0.2012	0.2536	0.2667	0.2257	0.7954	0.0208	0.0358	0.0	0.0	0.4346	0.725
1.2718	6.0	600	1.2374	0.1656	0.2413	0.1854	0.1213	0.0965	0.6428	0.0467	0.2083	0.2438	0.1963	0.2155	0.7912	0.007	0.0121	0.0	0.0	0.4897	0.7193
1.1692	7.0	700	1.2744	0.188	0.276	0.2086	0.0	0.1374	0.6403	0.0565	0.2173	0.2364	0.0	0.2066	0.7565	0.0238	0.0304	0.0	0.0	0.5402	0.6789
1.1205	8.0	800	1.1032	0.2248	0.3159	0.2623	0.1032	0.1919	0.6805	0.0781	0.264	0.285	0.2481	0.2604	0.785	0.0718	0.1358	0.0	0.0	0.6025	0.7191
1.1331	9.0	900	1.1086	0.2243	0.3351	0.2575	0.0355	0.1898	0.6761	0.0903	0.3044	0.331	0.2704	0.3025	0.7918	0.0842	0.2725	0.0002	0.0007	0.5884	0.7199
1.0573	10.0	1000	1.0497	0.2609	0.3848	0.2961	0.1714	0.2329	0.7023	0.1019	0.3431	0.3617	0.2556	0.331	0.7879	0.1495	0.3654	0.0084	0.0095	0.6248	0.7104
1.2048	11.0	1100	1.0155	0.3063	0.4403	0.3541	0.2089	0.2905	0.6859	0.1203	0.4108	0.4412	0.2333	0.4181	0.7663	0.2528	0.5717	0.0443	0.0486	0.6217	0.7033
1.0135	12.0	1200	0.9896	0.3031	0.4506	0.3501	0.3099	0.2756	0.6828	0.1195	0.4114	0.4319	0.3259	0.4	0.765	0.2659	0.5738	0.0333	0.0338	0.6101	0.6881
1.0212	13.0	1300	1.0943	0.3275	0.4723	0.3971	0.2373	0.3166	0.638	0.1258	0.444	0.4607	0.2481	0.4422	0.7229	0.3622	0.6779	0.0289	0.0358	0.5915	0.6684
0.9173	14.0	1400	0.9510	0.3389	0.4909	0.4013	0.2408	0.3205	0.6806	0.1251	0.4614	0.4832	0.363	0.4563	0.7683	0.359	0.7088	0.0406	0.0385	0.617	0.7023
0.8862	15.0	1500	0.9053	0.3771	0.5484	0.438	0.1616	0.3555	0.7143	0.1432	0.4874	0.5095	0.3148	0.4825	0.7882	0.4009	0.7129	0.0861	0.0953	0.6442	0.7203
0.9012	16.0	1600	0.9406	0.3889	0.5543	0.4661	0.185	0.3682	0.7036	0.1338	0.4758	0.4894	0.2704	0.466	0.7752	0.4708	0.6983	0.0509	0.0561	0.6451	0.7137
0.864	17.0	1700	0.8412	0.3906	0.5531	0.4728	0.2214	0.3729	0.7195	0.141	0.4916	0.514	0.2407	0.4925	0.7938	0.4593	0.7454	0.055	0.0608	0.6574	0.7357
0.7657	18.0	1800	0.8316	0.4046	0.5714	0.4843	0.1751	0.3865	0.7221	0.1539	0.5089	0.5296	0.3444	0.5031	0.7951	0.459	0.7546	0.0935	0.1054	0.6612	0.7287
0.8597	19.0	1900	0.8379	0.4105	0.5831	0.4949	0.2155	0.3935	0.7089	0.1394	0.4875	0.5033	0.2222	0.4807	0.7801	0.5315	0.7371	0.0509	0.0541	0.649	0.7188
0.7747	20.0	2000	0.8280	0.4182	0.5924	0.5143	0.2735	0.3993	0.6977	0.1392	0.4881	0.5025	0.2741	0.4789	0.7748	0.5537	0.735	0.0585	0.0601	0.6424	0.7125
0.7621	21.0	2100	0.7842	0.4345	0.6189	0.5267	0.1441	0.4145	0.7219	0.1461	0.5071	0.5213	0.3333	0.496	0.7915	0.556	0.7404	0.0915	0.1	0.6559	0.7234
0.7609	22.0	2200	0.7726	0.4249	0.5906	0.5262	0.1769	0.4043	0.7519	0.1291	0.4878	0.5039	0.3296	0.4788	0.8114	0.5697	0.7404	0.025	0.0223	0.68	0.749
0.7386	23.0	2300	0.7575	0.4478	0.6312	0.5444	0.3547	0.4234	0.7451	0.1477	0.5171	0.5308	0.3778	0.5035	0.8108	0.5688	0.7392	0.0972	0.1074	0.6773	0.7457
0.8086	24.0	2400	0.8128	0.4589	0.6456	0.5643	0.3348	0.4367	0.7173	0.1552	0.5206	0.5346	0.3407	0.5069	0.7863	0.5936	0.7433	0.1297	0.1419	0.6533	0.7186
0.7901	25.0	2500	0.7641	0.4724	0.6601	0.5711	0.2783	0.4507	0.7255	0.1638	0.54	0.5534	0.2926	0.5284	0.7895	0.5919	0.7571	0.1618	0.1777	0.6636	0.7254
0.7292	26.0	2600	0.7594	0.4769	0.6881	0.5651	0.3251	0.458	0.7166	0.1577	0.5427	0.555	0.3333	0.529	0.7918	0.5987	0.7479	0.1743	0.1899	0.6576	0.7273
0.6989	27.0	2700	0.7371	0.5386	0.7406	0.6675	0.3238	0.5192	0.7394	0.1833	0.6022	0.6144	0.3222	0.5895	0.8033	0.644	0.7683	0.2977	0.3372	0.6741	0.7377
0.7199	28.0	2800	0.6863	0.5527	0.7655	0.6766	0.3825	0.5321	0.7545	0.1845	0.6204	0.633	0.3889	0.6111	0.8141	0.6372	0.7758	0.3264	0.3696	0.6944	0.7537
0.6656	29.0	2900	0.6749	0.5543	0.7595	0.6732	0.3931	0.5343	0.7568	0.1765	0.608	0.6206	0.3963	0.5957	0.8183	0.6634	0.7704	0.3048	0.3385	0.6946	0.7529
0.6462	30.0	3000	0.6701	0.5793	0.8205	0.6987	0.3248	0.5603	0.7541	0.1903	0.6341	0.6478	0.3333	0.6225	0.8196	0.6641	0.7571	0.3821	0.4345	0.6916	0.752
0.6706	31.0	3100	0.6555	0.6146	0.8476	0.7482	0.364	0.5938	0.7543	0.2091	0.6742	0.6883	0.3704	0.6637	0.8193	0.6567	0.7583	0.4974	0.5547	0.6897	0.752
0.6464	32.0	3200	0.6445	0.6235	0.8621	0.7518	0.4682	0.6042	0.7563	0.207	0.6797	0.6951	0.4889	0.6696	0.819	0.6768	0.7713	0.4988	0.5601	0.6947	0.7539
0.5998	33.0	3300	0.6266	0.6292	0.8724	0.7738	0.4007	0.6132	0.7534	0.2113	0.6863	0.7008	0.4148	0.6787	0.818	0.6862	0.7758	0.5049	0.5682	0.6965	0.7584
0.6153	34.0	3400	0.6388	0.6202	0.8619	0.7475	0.417	0.6012	0.7515	0.2027	0.6755	0.6903	0.4185	0.6657	0.8167	0.6957	0.7804	0.4766	0.5392	0.6884	0.7512
0.5857	35.0	3500	0.6126	0.6426	0.8968	0.797	0.4822	0.6215	0.7521	0.218	0.6984	0.7132	0.4963	0.6895	0.8111	0.6907	0.7721	0.5477	0.6169	0.6894	0.7506
0.5625	36.0	3600	0.6348	0.6329	0.8926	0.7794	0.4634	0.6143	0.7436	0.2184	0.6909	0.7051	0.4815	0.6816	0.8082	0.6803	0.7679	0.5335	0.6	0.685	0.7475
0.5969	37.0	3700	0.6090	0.6643	0.9033	0.8282	0.4425	0.6453	0.7616	0.2295	0.7186	0.7334	0.4667	0.7117	0.8193	0.6971	0.7812	0.5944	0.6595	0.7013	0.7596
0.5882	38.0	3800	0.6197	0.6534	0.9061	0.811	0.4838	0.6356	0.7542	0.2241	0.7077	0.722	0.5	0.7002	0.8105	0.6798	0.7638	0.5863	0.6514	0.6941	0.7508
0.5455	39.0	3900	0.5811	0.6769	0.9265	0.8258	0.471	0.6573	0.7649	0.2312	0.7309	0.7461	0.4815	0.7225	0.8186	0.7126	0.7887	0.6156	0.6946	0.7023	0.7551
0.5654	40.0	4000	0.6227	0.662	0.9225	0.7937	0.4556	0.6472	0.749	0.2297	0.723	0.7351	0.4593	0.7155	0.8078	0.7022	0.7858	0.5892	0.6676	0.6945	0.7518
0.5414	41.0	4100	0.6159	0.6743	0.9156	0.813	0.4628	0.6597	0.753	0.234	0.7353	0.7477	0.4667	0.7275	0.8144	0.6968	0.7854	0.6274	0.7014	0.6987	0.7564
0.5337	42.0	4200	0.5916	0.6724	0.9252	0.8298	0.4644	0.656	0.7617	0.2346	0.7323	0.7456	0.463	0.7248	0.8183	0.6908	0.7758	0.6232	0.702	0.7034	0.759
0.5356	43.0	4300	0.6240	0.6713	0.9148	0.8097	0.5307	0.6517	0.767	0.2322	0.7304	0.7443	0.5333	0.7216	0.8225	0.6926	0.7825	0.6165	0.6899	0.7047	0.7605
0.5438	44.0	4400	0.6095	0.677	0.9178	0.8307	0.5792	0.6598	0.7614	0.235	0.7335	0.7477	0.5815	0.7264	0.8176	0.6938	0.7796	0.6331	0.7041	0.7042	0.7594
0.5397	45.0	4500	0.5821	0.6838	0.9288	0.8383	0.5075	0.6635	0.7693	0.2354	0.7406	0.7541	0.5185	0.7323	0.8232	0.7023	0.7883	0.6405	0.7108	0.7087	0.7633
0.4965	46.0	4600	0.5899	0.6809	0.9232	0.8247	0.4549	0.6644	0.7616	0.2351	0.7373	0.7505	0.4741	0.7294	0.8203	0.6995	0.7833	0.6402	0.7068	0.7031	0.7613
0.5718	47.0	4700	0.5733	0.6856	0.9248	0.8121	0.4957	0.666	0.7741	0.2389	0.7435	0.7561	0.5074	0.7326	0.8301	0.7131	0.7942	0.6321	0.7074	0.7117	0.7666
0.5055	48.0	4800	0.5774	0.6854	0.9236	0.8323	0.4892	0.6675	0.7673	0.2377	0.7399	0.7522	0.5037	0.7304	0.8196	0.7113	0.7887	0.6402	0.7088	0.7048	0.7592
0.5249	49.0	4900	0.5793	0.6855	0.9241	0.8387	0.506	0.6677	0.7659	0.2398	0.7394	0.7541	0.5185	0.7315	0.8196	0.7064	0.7862	0.6437	0.7176	0.7065	0.7584
0.5299	50.0	5000	0.5700	0.6888	0.9243	0.8388	0.4921	0.671	0.7666	0.2399	0.7411	0.7562	0.5037	0.7335	0.8235	0.7108	0.7892	0.648	0.7182	0.7075	0.7611
0.5028	51.0	5100	0.5649	0.6917	0.9274	0.844	0.5059	0.6739	0.7707	0.2404	0.7468	0.7606	0.5296	0.7371	0.8297	0.7104	0.7892	0.6552	0.725	0.7095	0.7678
0.5179	52.0	5200	0.5713	0.687	0.9277	0.8289	0.5104	0.6692	0.7662	0.2388	0.7419	0.7554	0.5222	0.7331	0.8216	0.7107	0.7921	0.6438	0.7128	0.7064	0.7611
0.5294	53.0	5300	0.5721	0.6889	0.9253	0.8264	0.4982	0.672	0.7665	0.2405	0.7444	0.7575	0.5111	0.7357	0.8225	0.7144	0.7946	0.6447	0.7155	0.7076	0.7625
0.528	54.0	5400	0.5722	0.6836	0.9247	0.8239	0.4957	0.6674	0.7658	0.2388	0.7388	0.754	0.5074	0.7324	0.8212	0.71	0.7937	0.6335	0.7068	0.7073	0.7615
0.498	55.0	5500	0.5722	0.6844	0.926	0.8314	0.4957	0.6676	0.7653	0.2393	0.7405	0.7541	0.5074	0.7317	0.8232	0.7128	0.7929	0.634	0.7074	0.7065	0.7619
0.5066	56.0	5600	0.5726	0.6853	0.9229	0.8279	0.4957	0.6691	0.7678	0.2396	0.741	0.7541	0.5074	0.7317	0.8242	0.7145	0.7929	0.6348	0.7068	0.7066	0.7627
0.4986	57.0	5700	0.5714	0.6872	0.9231	0.8224	0.4957	0.6713	0.7657	0.2396	0.7422	0.7552	0.5074	0.733	0.8235	0.7165	0.7937	0.6373	0.7095	0.7077	0.7625
0.4709	58.0	5800	0.5714	0.6869	0.9233	0.8248	0.5044	0.6709	0.7655	0.2394	0.7421	0.7552	0.5185	0.7331	0.8229	0.7157	0.7929	0.6374	0.7101	0.7075	0.7625
0.4974	59.0	5900	0.5714	0.6872	0.9234	0.8248	0.5069	0.6712	0.7654	0.2394	0.7422	0.7553	0.5185	0.733	0.8229	0.7156	0.7929	0.6388	0.7108	0.7073	0.7621
0.4991	60.0	6000	0.5714	0.6874	0.9234	0.8248	0.5069	0.6714	0.7654	0.2397	0.7424	0.7555	0.5185	0.7332	0.8229	0.7156	0.7929	0.6392	0.7115	0.7073	0.7621

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 2.19.2
• Tokenizers 0.20.0