chickens-60-epoch-1000-images-aug

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

• Loss: 0.2851
• Map: 0.8003
• Map 50: 0.9645
• Map 75: 0.9196
• Map Small: 0.2246
• Map Medium: 0.7999
• Map Large: 0.8772
• Mar 1: 0.3086
• Mar 10: 0.8346
• Mar 100: 0.8384
• Mar Small: 0.3614
• Mar Medium: 0.8496
• Mar Large: 0.918
• Map Chicken: 0.8072
• Mar 100 Chicken: 0.8413
• Map Duck: 0.7689
• Mar 100 Duck: 0.799
• Map Plant: 0.8248
• Mar 100 Plant: 0.8749

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: Use 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: 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
1.4811	1.0	500	1.3713	0.0772	0.1135	0.0875	0.004	0.0389	0.3534	0.0527	0.1857	0.2513	0.0875	0.2276	0.7937	0.0069	0.0062	0.0	0.0	0.2247	0.7476
1.093	2.0	1000	1.1411	0.1941	0.2637	0.221	0.0042	0.134	0.6715	0.0688	0.2331	0.2586	0.1167	0.23	0.8259	0.0	0.0	0.0	0.0	0.5822	0.7758
1.06	3.0	1500	1.5299	0.1865	0.2571	0.2063	0.0172	0.1497	0.6155	0.0695	0.2154	0.2187	0.0583	0.199	0.6971	0.0	0.0	0.0	0.0	0.5595	0.6562
0.866	4.0	2000	1.0132	0.2298	0.307	0.2534	0.0251	0.1936	0.7269	0.0897	0.2671	0.2739	0.0688	0.2449	0.8172	0.0329	0.0613	0.0	0.0	0.6564	0.7602
0.8301	5.0	2500	0.9037	0.2895	0.3972	0.3375	0.0486	0.25	0.7397	0.1158	0.3495	0.3568	0.1063	0.3238	0.8305	0.2025	0.3062	0.0	0.0	0.6662	0.7643
0.8386	6.0	3000	0.9659	0.3229	0.4625	0.3953	0.012	0.2944	0.7146	0.1161	0.3938	0.3977	0.0333	0.3785	0.7837	0.3161	0.4693	0.0	0.0	0.6527	0.7236
0.9838	7.0	3500	0.7706	0.3831	0.5318	0.4649	0.0314	0.3525	0.7573	0.1347	0.4783	0.4829	0.0854	0.4636	0.8234	0.4609	0.6889	0.0	0.0	0.6886	0.7599
0.8051	8.0	4000	0.7424	0.3909	0.5361	0.4744	0.0298	0.362	0.7506	0.1396	0.489	0.4921	0.1063	0.4762	0.8222	0.4864	0.712	0.0	0.0	0.6864	0.7643
0.7114	9.0	4500	0.6860	0.425	0.5595	0.5062	0.0656	0.4023	0.7822	0.1429	0.5123	0.5168	0.1104	0.5014	0.849	0.5552	0.7609	0.0	0.0	0.7197	0.7896
0.8088	10.0	5000	0.6922	0.4107	0.5706	0.5008	0.0425	0.3953	0.7653	0.1375	0.4929	0.4981	0.1333	0.4833	0.8272	0.527	0.7218	0.0	0.0	0.705	0.7726
0.7049	11.0	5500	0.6989	0.4204	0.5653	0.5121	0.0859	0.4022	0.7752	0.1439	0.4952	0.499	0.1458	0.4863	0.8293	0.5406	0.7182	0.0	0.0	0.7205	0.7787
0.7244	12.0	6000	0.6311	0.4276	0.584	0.5016	0.0749	0.4104	0.7945	0.1445	0.4989	0.5035	0.1542	0.4876	0.8515	0.5447	0.7133	0.0	0.0	0.7382	0.7971
0.683	13.0	6500	0.6244	0.4371	0.5962	0.5288	0.0877	0.4149	0.7935	0.1447	0.5002	0.5031	0.1312	0.486	0.8427	0.5784	0.7244	0.0	0.0	0.7329	0.7847
0.6541	14.0	7000	0.5543	0.4719	0.6191	0.5555	0.0712	0.4525	0.8231	0.1494	0.5195	0.5249	0.1896	0.5025	0.8715	0.6567	0.7644	0.0	0.0	0.759	0.8104
0.6219	15.0	7500	0.5368	0.4754	0.6197	0.5553	0.0764	0.4584	0.825	0.1528	0.5216	0.5265	0.1688	0.5038	0.8711	0.6691	0.7724	0.0	0.0	0.7571	0.8072
0.5842	16.0	8000	0.5325	0.4778	0.6269	0.5668	0.1147	0.4558	0.8015	0.1501	0.5178	0.5218	0.1604	0.5071	0.8556	0.6922	0.7636	0.0	0.0	0.7412	0.8017
0.5704	17.0	8500	0.5437	0.5192	0.6982	0.6149	0.0616	0.5014	0.8084	0.1798	0.558	0.5618	0.1521	0.5445	0.8644	0.6772	0.7449	0.1347	0.1412	0.7456	0.7994
0.5683	18.0	9000	0.5068	0.6324	0.8451	0.7659	0.0963	0.6253	0.8208	0.225	0.6739	0.6793	0.175	0.6808	0.8628	0.6996	0.7573	0.4404	0.4753	0.7573	0.8052
0.6402	19.0	9500	0.4682	0.6741	0.8823	0.8298	0.1357	0.6748	0.8274	0.2516	0.7135	0.7185	0.2104	0.7246	0.8728	0.7195	0.7698	0.5335	0.567	0.7691	0.8187
0.5664	20.0	10000	0.4793	0.6841	0.9057	0.8277	0.135	0.6878	0.8164	0.2585	0.7299	0.7341	0.2396	0.7463	0.8649	0.7325	0.7853	0.5558	0.5979	0.7638	0.819
0.4411	21.0	10500	0.4448	0.7042	0.932	0.8592	0.1098	0.7039	0.8287	0.2789	0.7527	0.7568	0.1718	0.7703	0.8749	0.7128	0.7658	0.6338	0.6845	0.766	0.8202
0.6106	22.0	11000	0.4142	0.7307	0.9307	0.8797	0.0773	0.735	0.8381	0.2841	0.7736	0.7783	0.2062	0.7946	0.8866	0.7379	0.7853	0.6726	0.7134	0.7817	0.836
0.5243	23.0	11500	0.4353	0.7183	0.9406	0.86	0.0901	0.7236	0.8416	0.2827	0.7615	0.767	0.1973	0.779	0.8879	0.7338	0.7827	0.6385	0.6845	0.7827	0.8337
0.5184	24.0	12000	0.4077	0.7097	0.9464	0.854	0.1197	0.7156	0.8335	0.2741	0.757	0.7607	0.2553	0.7738	0.8828	0.7126	0.7667	0.6338	0.6784	0.7828	0.8372
0.4849	25.0	12500	0.4043	0.7096	0.949	0.8366	0.1234	0.7084	0.8412	0.2739	0.7538	0.7611	0.258	0.7703	0.8891	0.7483	0.7902	0.596	0.6546	0.7843	0.8383
0.5022	26.0	13000	0.3884	0.7394	0.9528	0.8847	0.1472	0.7337	0.8473	0.2918	0.7816	0.7876	0.2549	0.7888	0.8971	0.7466	0.7871	0.688	0.7402	0.7838	0.8354
0.521	27.0	13500	0.4197	0.7177	0.9434	0.8715	0.132	0.7168	0.8353	0.2879	0.7639	0.7697	0.2623	0.7777	0.8799	0.7073	0.7649	0.6685	0.7165	0.7771	0.8277
0.5433	28.0	14000	0.3886	0.7454	0.9508	0.8823	0.2083	0.7406	0.8448	0.292	0.7833	0.789	0.3064	0.7952	0.8845	0.7573	0.8004	0.6941	0.733	0.785	0.8334
0.3889	29.0	14500	0.3713	0.7492	0.9553	0.8998	0.2224	0.7492	0.8468	0.2891	0.7873	0.7936	0.3112	0.8026	0.8921	0.7677	0.8053	0.6849	0.7299	0.7951	0.8455
0.5103	30.0	15000	0.3556	0.7584	0.9576	0.9014	0.219	0.7509	0.8517	0.2939	0.7954	0.8015	0.3089	0.8078	0.8979	0.7654	0.8022	0.7165	0.7557	0.7934	0.8467
0.4458	31.0	15500	0.3681	0.7355	0.9518	0.8831	0.1416	0.7311	0.8569	0.292	0.7773	0.7814	0.2358	0.7892	0.9008	0.734	0.7742	0.6771	0.7247	0.7955	0.8452
0.4369	32.0	16000	0.3523	0.7495	0.9499	0.8877	0.142	0.7447	0.8514	0.2935	0.7877	0.7923	0.2589	0.8026	0.895	0.765	0.8027	0.6903	0.7309	0.7932	0.8432
0.447	33.0	16500	0.3665	0.7448	0.954	0.8912	0.1577	0.7412	0.8505	0.2922	0.7844	0.7879	0.2742	0.8005	0.8929	0.7453	0.7813	0.6933	0.7361	0.796	0.8464
0.4692	34.0	17000	0.3455	0.7589	0.954	0.899	0.1729	0.7459	0.863	0.2949	0.798	0.8038	0.3123	0.8068	0.9033	0.7709	0.8093	0.7058	0.7546	0.7999	0.8476
0.4272	35.0	17500	0.3381	0.767	0.9568	0.903	0.1734	0.7623	0.852	0.2948	0.802	0.8061	0.2907	0.8136	0.8962	0.7842	0.8204	0.7246	0.7546	0.7922	0.8432
0.4021	36.0	18000	0.3323	0.7686	0.9551	0.8938	0.1892	0.7621	0.8616	0.2969	0.8025	0.8067	0.3049	0.8122	0.9025	0.7776	0.8133	0.7245	0.7577	0.8038	0.849
0.4582	37.0	18500	0.3263	0.7732	0.9547	0.9023	0.1477	0.7748	0.8742	0.3015	0.8092	0.8132	0.2634	0.8251	0.9163	0.7729	0.812	0.7298	0.7619	0.817	0.8657
0.3992	38.0	19000	0.3207	0.7799	0.956	0.9064	0.1767	0.7823	0.873	0.3014	0.8168	0.8209	0.3028	0.8312	0.9172	0.7833	0.8218	0.7384	0.7732	0.818	0.8677
0.4286	39.0	19500	0.3194	0.7717	0.9567	0.8986	0.1626	0.7677	0.8779	0.3033	0.8101	0.8139	0.2835	0.82	0.918	0.7752	0.8204	0.7223	0.7598	0.8175	0.8614
0.4488	40.0	20000	0.3184	0.7718	0.9566	0.9047	0.1921	0.7702	0.8809	0.2999	0.8092	0.8141	0.3002	0.8238	0.9192	0.7776	0.8187	0.7156	0.7546	0.8223	0.8689
0.3763	41.0	20500	0.3055	0.7876	0.956	0.9186	0.1841	0.7824	0.8844	0.3033	0.8207	0.8254	0.3061	0.8339	0.9234	0.7973	0.8356	0.7394	0.7691	0.8262	0.8715
0.5658	42.0	21000	0.3014	0.791	0.9594	0.9167	0.2095	0.7911	0.8786	0.3032	0.8244	0.8299	0.3403	0.8403	0.9213	0.805	0.8404	0.7421	0.7742	0.8259	0.8749
0.4322	43.0	21500	0.2974	0.7974	0.9595	0.9169	0.1879	0.7927	0.8951	0.3078	0.8296	0.834	0.3085	0.8411	0.931	0.8005	0.8369	0.7586	0.7876	0.8331	0.8775
0.7057	44.0	22000	0.3092	0.7822	0.9563	0.9171	0.1985	0.7813	0.8688	0.3003	0.8165	0.821	0.3663	0.8292	0.9117	0.7941	0.8307	0.7348	0.766	0.8177	0.8663
0.4096	45.0	22500	0.2991	0.7899	0.9614	0.9121	0.2212	0.7852	0.8747	0.3031	0.8233	0.8286	0.3578	0.8351	0.9142	0.8016	0.8413	0.7502	0.7794	0.8179	0.8651
0.4854	46.0	23000	0.3003	0.7815	0.9595	0.9068	0.2042	0.7791	0.8747	0.3016	0.8164	0.8197	0.3258	0.8255	0.9163	0.7816	0.8231	0.746	0.7722	0.8169	0.8637
0.4257	47.0	23500	0.2951	0.792	0.9625	0.9172	0.2075	0.7855	0.8802	0.3067	0.8262	0.8309	0.3468	0.836	0.9197	0.7961	0.8338	0.7572	0.7907	0.8226	0.8683
0.4033	48.0	24000	0.2883	0.7988	0.9632	0.9194	0.2266	0.7984	0.8765	0.3082	0.8343	0.8382	0.3616	0.8477	0.9176	0.8069	0.8458	0.7649	0.7969	0.8246	0.872
0.4932	49.0	24500	0.3022	0.7844	0.9617	0.9101	0.2231	0.7765	0.8762	0.3007	0.8216	0.8252	0.3396	0.8308	0.9176	0.7882	0.8293	0.7472	0.7794	0.8177	0.8669
0.3758	50.0	25000	0.2959	0.7921	0.9609	0.9203	0.2432	0.7853	0.8779	0.3066	0.8273	0.8314	0.3655	0.8365	0.9197	0.7932	0.832	0.7619	0.7918	0.8212	0.8703
0.4397	51.0	25500	0.2871	0.7983	0.9609	0.9128	0.2145	0.7966	0.8832	0.3086	0.833	0.8374	0.3409	0.8478	0.9251	0.802	0.8369	0.7648	0.7969	0.828	0.8784
0.3917	52.0	26000	0.2907	0.7955	0.9645	0.9161	0.2316	0.7911	0.8796	0.308	0.8314	0.8352	0.375	0.8428	0.9192	0.7975	0.8356	0.7654	0.7969	0.8234	0.8732
0.3362	53.0	26500	0.2885	0.7989	0.9644	0.92	0.2324	0.7958	0.8789	0.3075	0.8338	0.8379	0.3769	0.8465	0.9201	0.8012	0.8382	0.7703	0.8	0.8253	0.8755
0.4004	54.0	27000	0.2869	0.7973	0.9644	0.9201	0.228	0.7957	0.8813	0.3069	0.8328	0.8368	0.3822	0.8456	0.9218	0.801	0.8373	0.7636	0.7948	0.8273	0.8781
0.406	55.0	27500	0.2871	0.8004	0.9645	0.9194	0.2283	0.7986	0.8788	0.3084	0.8343	0.8384	0.3822	0.8476	0.9205	0.8069	0.8404	0.7679	0.7979	0.8265	0.8769
0.3876	56.0	28000	0.2882	0.7985	0.9641	0.9197	0.2257	0.7974	0.8772	0.3084	0.834	0.838	0.3676	0.8474	0.918	0.8072	0.8436	0.7646	0.7969	0.8237	0.8735
0.3939	57.0	28500	0.2845	0.8024	0.9645	0.9195	0.2291	0.8014	0.8782	0.3093	0.8367	0.8405	0.3697	0.8511	0.9197	0.8102	0.8431	0.7709	0.8021	0.8262	0.8764
0.4218	58.0	29000	0.2852	0.8	0.9646	0.9196	0.2254	0.7993	0.8774	0.3085	0.8346	0.8384	0.3634	0.8488	0.9188	0.8067	0.8413	0.7689	0.799	0.8245	0.8749
0.4046	59.0	29500	0.2851	0.8008	0.9645	0.9196	0.2283	0.8002	0.878	0.3087	0.835	0.8388	0.3655	0.8497	0.9188	0.8079	0.8418	0.7689	0.799	0.8256	0.8758
0.4504	60.0	30000	0.2851	0.8003	0.9645	0.9196	0.2246	0.7999	0.8772	0.3086	0.8346	0.8384	0.3614	0.8496	0.918	0.8072	0.8413	0.7689	0.799	0.8248	0.8749

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.0+cu121
• Datasets 2.19.2
• Tokenizers 0.20.1