detr-arrows

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: 1.0051
• Map: 0.0157
• Map 50: 0.0235
• Map 75: 0.017
• Map Small: 0.0157
• Map Medium: -1.0
• Map Large: -1.0
• Mar 1: 0.0854
• Mar 10: 0.0917
• Mar 100: 0.2917
• Mar Small: 0.2917
• Mar Medium: -1.0
• Mar Large: -1.0
• Map Left: 0.0
• Mar 100 Left: 0.0
• Map Right: -1.0
• Mar 100 Right: -1.0
• Map Up: 0.0385
• Mar 100 Up: 0.2667
• Map Down: 0.0146
• Mar 100 Down: 0.8
• Map ?: 0.0099
• Mar 100 ?: 0.1

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: 3e-05
• train_batch_size: 8
• 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 Left	Mar 100 Left	Map Right	Mar 100 Right	Map Up	Mar 100 Up	Map Down	Mar 100 Down	Map ?	Mar 100 ?
No log	1.0	8	2.9834	0.0001	0.0014	0.0	0.0002	-1.0	-1.0	0.0	0.0	0.025	0.025	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0006	0.1	0.0	0.0
No log	2.0	16	2.3564	0.0005	0.0023	0.0	0.0007	-1.0	-1.0	0.0	0.0	0.05	0.05	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0019	0.2	0.0	0.0
No log	3.0	24	2.1618	0.001	0.0025	0.0	0.0012	-1.0	-1.0	0.0	0.0	0.1	0.1	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.004	0.4	0.0	0.0
No log	4.0	32	2.0074	0.0005	0.0027	0.0	0.0007	-1.0	-1.0	0.0	0.0	0.05	0.05	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0022	0.2	0.0	0.0
No log	5.0	40	1.7452	0.0008	0.003	0.0	0.0009	-1.0	-1.0	0.0	0.0	0.0812	0.0812	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0028	0.3	0.0003	0.025
No log	6.0	48	1.6746	0.0009	0.0024	0.0	0.0011	-1.0	-1.0	0.0	0.0	0.1	0.1	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0038	0.4	0.0	0.0
No log	7.0	56	1.5161	0.0014	0.0023	0.0023	0.0016	-1.0	-1.0	0.0	0.0	0.15	0.15	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0054	0.6	0.0	0.0
No log	8.0	64	1.5721	0.0009	0.0017	0.0	0.001	-1.0	-1.0	0.0	0.0	0.125	0.125	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0034	0.5	0.0	0.0
No log	9.0	72	1.4536	0.0014	0.0021	0.0021	0.0021	-1.0	-1.0	0.0	0.0	0.175	0.175	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0056	0.7	0.0	0.0
No log	10.0	80	1.2638	0.0013	0.0023	0.0014	0.0019	-1.0	-1.0	0.0	0.0	0.175	0.175	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0052	0.7	0.0	0.0
No log	11.0	88	1.2646	0.0015	0.0025	0.0014	0.0025	-1.0	-1.0	0.0	0.0	0.175	0.175	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0062	0.7	0.0	0.0
No log	12.0	96	1.2239	0.0023	0.003	0.003	0.0038	-1.0	-1.0	0.0	0.0	0.2	0.2	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0092	0.8	0.0	0.0
No log	13.0	104	1.1191	0.0023	0.0057	0.0025	0.0035	-1.0	-1.0	0.0	0.0125	0.1875	0.1875	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0068	0.7	0.0026	0.05
No log	14.0	112	1.1491	0.0029	0.0054	0.002	0.0038	-1.0	-1.0	0.0312	0.0312	0.1813	0.1813	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0	0.0	0.0047	0.6	0.0068	0.125
No log	15.0	120	1.1062	0.02	0.0349	0.0297	0.0287	-1.0	-1.0	0.05	0.075	0.25	0.25	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0673	0.2	0.0053	0.7	0.0074	0.1
No log	16.0	128	1.0963	0.0705	0.0925	0.086	0.0705	-1.0	-1.0	0.0854	0.0854	0.2354	0.2354	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.2693	0.2667	0.0058	0.6	0.007	0.075
No log	17.0	136	1.0369	0.0349	0.051	0.0434	0.0349	-1.0	-1.0	0.075	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.1234	0.2667	0.0049	0.8	0.0114	0.1
No log	18.0	144	1.0721	0.0324	0.05	0.0435	0.0324	-1.0	-1.0	0.0771	0.0771	0.2521	0.2521	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.1178	0.2333	0.004	0.7	0.0077	0.075
No log	19.0	152	1.0081	0.0623	0.0922	0.0858	0.0625	-1.0	-1.0	0.0646	0.0833	0.2833	0.2833	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.2356	0.2333	0.0051	0.8	0.0085	0.1
No log	20.0	160	1.0390	0.0367	0.0504	0.044	0.0367	-1.0	-1.0	0.0729	0.0854	0.2604	0.2604	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.1347	0.2667	0.0053	0.7	0.0071	0.075
No log	21.0	168	1.0262	0.0255	0.0371	0.0299	0.0255	-1.0	-1.0	0.0729	0.0854	0.2604	0.2604	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0898	0.2667	0.0052	0.7	0.0068	0.075
No log	22.0	176	0.9972	0.0134	0.0223	0.0159	0.0134	-1.0	-1.0	0.0646	0.0833	0.2833	0.2833	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0393	0.2333	0.0058	0.8	0.0086	0.1
No log	23.0	184	1.0195	0.0135	0.0204	0.014	0.0135	-1.0	-1.0	0.0729	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.0061	0.8	0.0092	0.1
No log	24.0	192	1.0240	0.0145	0.023	0.0139	0.0145	-1.0	-1.0	0.0792	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.0101	0.8	0.0095	0.1
No log	25.0	200	1.0151	0.0157	0.024	0.0176	0.0157	-1.0	-1.0	0.0854	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.0147	0.8	0.0096	0.1
No log	26.0	208	1.0143	0.0153	0.0233	0.0168	0.0153	-1.0	-1.0	0.0854	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.013	0.8	0.0099	0.1
No log	27.0	216	1.0092	0.0157	0.0235	0.017	0.0157	-1.0	-1.0	0.0854	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.0146	0.8	0.0099	0.1
No log	28.0	224	1.0051	0.0157	0.0235	0.017	0.0157	-1.0	-1.0	0.0854	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.0146	0.8	0.0099	0.1
No log	29.0	232	1.0046	0.0157	0.0235	0.017	0.0157	-1.0	-1.0	0.0854	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.0146	0.8	0.0099	0.1
No log	30.0	240	1.0051	0.0157	0.0235	0.017	0.0157	-1.0	-1.0	0.0854	0.0917	0.2917	0.2917	-1.0	-1.0	0.0	0.0	-1.0	-1.0	0.0385	0.2667	0.0146	0.8	0.0099	0.1

Framework versions

• Transformers 4.48.0.dev0
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0