rtdetr-r50-cppe5-finetune-use_focal-False

qubvel-hf/rtdetr-r50-cppe5-finetune-use_focal-False

rtdetr-r50-cppe5-finetune-use_focal-False is a machine learning model from qubvel-hf. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

10% of all-time downloads

All-time downloads

142

Public

Parameters

42.9M

343 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors172 MB · 100%

At a glance

License: apache-2.0
Model type: rt_detr
Access: Public
Created: Aug 12, 2024
Updated: Aug 12, 2024
SHA: ce501b0b

Base models

PekingU/rtdetr_r50vd_coco_o365

Model card

From the Hugging Face model README

rtdetr-r50-cppe5-finetune-use_focal-False

This model is a fine-tuned version of PekingU/rtdetr_r50vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 10.2265
Map: 0.3751
Map 50: 0.5356
Map 75: 0.4176
Map Small: 0.5517
Map Medium: 0.3568
Map Large: 0.5283
Mar 1: 0.329
Mar 10: 0.6005
Mar 100: 0.6462
Mar Small: 0.5882
Mar Medium: 0.5601
Mar Large: 0.7161
Map Coverall: 0.3237
Mar 100 Coverall: 0.8564
Map Face Shield: 0.4291
Mar 100 Face Shield: 0.8467
Map Gloves: 0.6116
Mar 100 Gloves: 0.7638
Map Goggles: 0.5111
Mar 100 Goggles: 0.7643
Map Mask: 0.0
Mar 100 Mask: 0.0

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 300
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 Coverall	Mar 100 Coverall	Map Face Shield	Mar 100 Face Shield	Map Gloves	Mar 100 Gloves	Map Goggles	Mar 100 Goggles
No log	1.0	106	48.0056	0.0096	0.0182	0.0078	0.0004	0.0011	0.0141	0.0279	0.0877	0.1286	0.0364	0.0457	0.2074	0.0454	0.3986	0.0008	0.0467	0.0014	0.0836	0.0002	0.1141
No log	2.0	212	24.1177	0.0533	0.0967	0.0483	0.0429	0.0161	0.0799	0.0924	0.2249	0.2994	0.1024	0.1912	0.4803	0.1701	0.63	0.0057	0.3213	0.07	0.3301	0.0207	0.2156
No log	3.0	318	19.6837	0.0909	0.1714	0.0842	0.1028	0.0561	0.1558	0.168	0.3546	0.4261	0.2402	0.3345	0.5601	0.1759	0.6977	0.0739	0.508	0.1338	0.5059	0.0708	0.4187
No log	4.0	424	15.9325	0.1186	0.2136	0.1107	0.0801	0.0859	0.2414	0.1884	0.3751	0.4496	0.262	0.3839	0.5603	0.2001	0.7092	0.0708	0.5373	0.1727	0.5123	0.1493	0.4891
46.7509	5.0	530	15.7219	0.1652	0.2835	0.1715	0.1035	0.13	0.2798	0.2215	0.3943	0.4835	0.3143	0.4257	0.58	0.2815	0.7286	0.1155	0.616	0.2126	0.5466	0.2163	0.5266
46.7509	6.0	636	15.2971	0.1521	0.2608	0.1585	0.1615	0.107	0.3026	0.2048	0.3936	0.4832	0.3281	0.4196	0.5982	0.2162	0.7461	0.1248	0.608	0.2377	0.5525	0.1817	0.5094
46.7509	7.0	742	14.7975	0.1739	0.3033	0.1663	0.1774	0.1249	0.2946	0.22	0.4152	0.4941	0.3349	0.4265	0.6162	0.2714	0.7359	0.1555	0.64	0.2563	0.5648	0.1865	0.5297
46.7509	8.0	848	14.0780	0.173	0.3097	0.17	0.1574	0.1366	0.3098	0.2273	0.4161	0.491	0.3402	0.4332	0.6074	0.2117	0.7401	0.1694	0.6413	0.2664	0.5626	0.2175	0.5109
46.7509	9.0	954	14.7545	0.1838	0.3225	0.1884	0.1876	0.1506	0.3254	0.2366	0.4295	0.5001	0.3676	0.4306	0.6197	0.2092	0.7424	0.1855	0.648	0.2837	0.563	0.2407	0.5469
15.0328	10.0	1060	14.8555	0.1901	0.3198	0.1947	0.1801	0.1621	0.3094	0.2376	0.4254	0.4966	0.3667	0.4266	0.6081	0.2333	0.7355	0.2159	0.6267	0.274	0.5676	0.2271	0.5531
15.0328	11.0	1166	14.5398	0.2122	0.3554	0.2155	0.1985	0.1684	0.3698	0.2458	0.4265	0.5005	0.3759	0.4312	0.6075	0.3117	0.7475	0.2296	0.6307	0.2706	0.5667	0.2489	0.5578
15.0328	12.0	1272	13.9358	0.2154	0.3669	0.2171	0.1878	0.1602	0.3618	0.2421	0.4238	0.5003	0.3679	0.4206	0.6028	0.2657	0.7433	0.2847	0.6507	0.2909	0.5763	0.2358	0.5312
15.0328	13.0	1378	13.4203	0.2098	0.3538	0.2204	0.1888	0.1669	0.344	0.25	0.4176	0.5	0.3465	0.4387	0.6184	0.2666	0.7558	0.2454	0.6707	0.3017	0.5813	0.2355	0.4922
15.0328	14.0	1484	13.6884	0.1899	0.3258	0.1903	0.1759	0.169	0.3004	0.2455	0.4249	0.4988	0.3645	0.4277	0.6163	0.2431	0.7456	0.25	0.656	0.2834	0.5877	0.1732	0.5047
13.3953	15.0	1590	13.1320	0.1898	0.3258	0.192	0.1843	0.1431	0.3191	0.2492	0.4182	0.4935	0.362	0.4115	0.6128	0.2222	0.7401	0.2256	0.6347	0.2892	0.5694	0.2119	0.5234
13.3953	16.0	1696	12.9858	0.2019	0.3335	0.2047	0.1801	0.1637	0.3552	0.2554	0.4233	0.5005	0.3673	0.4268	0.6094	0.2745	0.7498	0.2364	0.644	0.2863	0.5712	0.2123	0.5375
13.3953	17.0	1802	12.9965	0.2013	0.3422	0.1992	0.1809	0.1569	0.3458	0.248	0.4273	0.4914	0.3632	0.4103	0.6139	0.2725	0.7392	0.227	0.628	0.283	0.5712	0.2238	0.5188
13.3953	18.0	1908	12.9245	0.1948	0.3346	0.1923	0.186	0.1505	0.3242	0.2386	0.4303	0.5049	0.393	0.4192	0.6206	0.2743	0.7392	0.219	0.644	0.2984	0.574	0.1824	0.5672
12.6173	19.0	2014	12.9508	0.209	0.3508	0.2072	0.1722	0.15	0.3759	0.2444	0.4307	0.5022	0.3574	0.4338	0.6149	0.2605	0.7456	0.2698	0.6533	0.2983	0.584	0.2166	0.5281
12.6173	20.0	2120	13.3318	0.2137	0.3577	0.223	0.1887	0.168	0.3762	0.2458	0.4233	0.4925	0.3564	0.4209	0.6174	0.2679	0.7382	0.2664	0.636	0.2961	0.5662	0.2382	0.5219
12.6173	21.0	2226	13.0245	0.2147	0.3564	0.2197	0.1747	0.1795	0.3656	0.2553	0.4266	0.4976	0.3564	0.4265	0.6323	0.2778	0.7387	0.2786	0.6533	0.299	0.5758	0.2182	0.5203
12.6173	22.0	2332	12.9212	0.2161	0.3697	0.2196	0.1798	0.1758	0.3509	0.2553	0.4358	0.5001	0.3676	0.4165	0.6155	0.2721	0.7355	0.2841	0.648	0.3008	0.5813	0.2235	0.5359
12.6173	23.0	2438	12.9598	0.2229	0.3751	0.2366	0.1823	0.177	0.3596	0.2533	0.4256	0.4976	0.3697	0.4193	0.6026	0.2901	0.7401	0.3074	0.6507	0.2954	0.5831	0.2218	0.5141
12.1016	24.0	2544	12.9207	0.2141	0.3637	0.2115	0.1954	0.1776	0.3447	0.2464	0.4278	0.4968	0.3665	0.4224	0.607	0.2536	0.7378	0.3149	0.6373	0.2955	0.5776	0.2064	0.5312
12.1016	25.0	2650	12.8912	0.2215	0.3703	0.2287	0.197	0.175	0.3619	0.2538	0.4312	0.4979	0.3657	0.4263	0.6035	0.2812	0.7465	0.3006	0.628	0.3082	0.5913	0.2173	0.5234
12.1016	26.0	2756	12.7256	0.2198	0.3724	0.2245	0.1926	0.1776	0.3522	0.2465	0.4341	0.5016	0.3742	0.4312	0.6073	0.2757	0.7479	0.2864	0.6333	0.3018	0.5831	0.2351	0.5437
12.1016	27.0	2862	12.7695	0.2259	0.3771	0.2409	0.1941	0.1854	0.3571	0.249	0.433	0.5027	0.3773	0.4412	0.591	0.2804	0.7447	0.3012	0.6333	0.3023	0.5872	0.2457	0.5484
12.1016	28.0	2968	12.8203	0.2231	0.3781	0.2333	0.1961	0.1837	0.3527	0.2492	0.4328	0.5001	0.3718	0.4331	0.607	0.2634	0.7456	0.3069	0.6347	0.3026	0.5872	0.2427	0.5328
11.5997	29.0	3074	12.7159	0.2205	0.3758	0.2306	0.194	0.177	0.3562	0.251	0.4322	0.5029	0.3882	0.4349	0.6057	0.258	0.7465	0.3073	0.6413	0.3037	0.5826	0.2336	0.5437
11.5997	30.0	3180	12.7243	0.2205	0.3735	0.2311	0.1938	0.1773	0.3609	0.2515	0.4326	0.5016	0.3869	0.4314	0.5982	0.2638	0.7438	0.2967	0.6347	0.3029	0.584	0.2393	0.5453

Framework versions

Transformers 4.45.0.dev0
Pytorch 2.4.0+cu121
Datasets 2.20.0
Tokenizers 0.19.1

Type: rt_detr
License: apache-2.0
Created: Aug 12, 2024
Updated: Aug 12, 2024