dfine-small-construction-ppe-v1

thalostech2025/dfine-small-construction-ppe-v1

dfine-small-construction-ppe-v1 is a object detection model from thalostech2025. Use it when you need objects located in an image. It is set up for transformers. The card lists the license as apache-2.0.

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

Downloads · 30 days

12% of all-time downloads

All-time downloads

398

Public

Parameters

10.3M

1.3 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors41.1 MB · 100%

At a glance

Task: Object Detection
Library: transformers
License: apache-2.0
Model type: d_fine
Access: Public
Created: Jun 29, 2026
Updated: Jun 29, 2026
SHA: 4931a416

Base models

ustc-community/dfine-small-coco

Model card

From the Hugging Face model README

dfine-small-construction-ppe-v1

This model is a fine-tuned version of ustc-community/dfine-small-coco on the thalostech2025/construction-ppe-equipment-v1 dataset. It achieves the following results on the evaluation set:

Loss: 2.6627
Map: 0.3579
Map 50: 0.4601
Map 75: 0.4002
Map Small: 0.1736
Map Medium: 0.2436
Map Large: 0.4745
Mar 1: 0.3584
Mar 10: 0.543
Mar 100: 0.5699
Mar Small: 0.4171
Mar Medium: 0.476
Mar Large: 0.7197
Map Person: 0.7402
Mar 100 Person: 0.8192
Map Hardhat: 0.5973
Mar 100 Hardhat: 0.7
Map No-hardhat: 0.1582
Mar 100 No-hardhat: 0.6222
Map Safety-vest: 0.7389
Mar 100 Safety-vest: 0.7948
Map No-safety-vest: 0.001
Mar 100 No-safety-vest: 0.2733
Map No-mask: 0.0
Mar 100 No-mask: 0.0214
Map No-goggles: 0.0
Mar 100 No-goggles: 0.0667
Map Gloves: 0.1501
Mar 100 Gloves: 0.4205
Map Safety-boots: 0.2746
Mar 100 Safety-boots: 0.4958
Map Excavator: 0.4878
Mar 100 Excavator: 0.8667
Map Dump-truck: 0.2657
Mar 100 Dump-truck: 0.8125
Map Wheel-loader: 0.8808
Mar 100 Wheel-loader: 0.9455

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 Person	Mar 100 Person	Map Hardhat	Mar 100 Hardhat	Map No-hardhat	Mar 100 No-hardhat	Map Safety-vest	Mar 100 Safety-vest	Map No-safety-vest	Mar 100 No-safety-vest	Mar 100 No-mask	Map No-goggles	Mar 100 No-goggles	Map Gloves	Mar 100 Gloves	Map Safety-boots	Mar 100 Safety-boots	Map Excavator	Mar 100 Excavator	Map Dump-truck	Mar 100 Dump-truck	Map Wheel-loader	Mar 100 Wheel-loader
No log	1.0	305	4.7930	0.1855	0.2742	0.2056	0.0526	0.1212	0.2218	0.27	0.3917	0.4416	0.1652	0.3143	0.546	0.5218	0.7423	0.3089	0.6361	0.0002	0.1778	0.5569	0.7134	0.0001	0.1533	0.0	0.0	0.0	0.0588	0.378	0.0266	0.1955	0.0225	0.6667	0.0307	0.7	0.7	0.9364
23.7150	2.0	610	4.1548	0.2186	0.3142	0.2439	0.0779	0.1365	0.2686	0.2542	0.4423	0.5005	0.3825	0.3748	0.6197	0.5383	0.7212	0.2043	0.676	0.0014	0.4667	0.6448	0.728	0.0001	0.2333	0.0	0.0	0.0	0.0543	0.497	0.0196	0.3968	0.2704	0.6667	0.1121	0.675	0.7785	0.9455
23.7150	3.0	915	3.8318	0.2425	0.3272	0.2749	0.0746	0.1746	0.3147	0.2862	0.503	0.5672	0.3837	0.446	0.7038	0.6169	0.7836	0.336	0.7097	0.0036	0.5889	0.6846	0.7711	0.0002	0.2933	0.0	0.0	0.0	0.0662	0.5284	0.0303	0.4415	0.1525	0.8667	0.1132	0.85	0.9064	0.9727
16.4356	4.0	1220	3.6373	0.2426	0.3284	0.2764	0.0907	0.1904	0.3053	0.286	0.5036	0.5815	0.3754	0.4541	0.736	0.6611	0.7976	0.4348	0.7186	0.0039	0.5889	0.7003	0.7795	0.0002	0.3467	0.0	0.0	0.0	0.0742	0.4914	0.0542	0.549	0.0726	0.9	0.1286	0.825	0.7806	0.9818
14.7326	5.0	1525	3.6059	0.2823	0.3548	0.3186	0.1088	0.1944	0.3472	0.3331	0.5098	0.5627	0.337	0.445	0.7217	0.6482	0.7762	0.3809	0.7075	0.0049	0.5889	0.6988	0.7718	0.0003	0.2867	0.0	0.0	0.0	0.0648	0.3799	0.047	0.5063	0.4286	0.9	0.2014	0.8625	0.9119	0.9727
14.7326	6.0	1830	3.4337	0.2678	0.3497	0.2965	0.1087	0.177	0.3483	0.2918	0.5307	0.5805	0.3134	0.483	0.7436	0.697	0.8062	0.454	0.7219	0.0129	0.6111	0.7203	0.7923	0.0005	0.32	0.0	0.0	0.0	0.0774	0.4362	0.0622	0.5333	0.2258	0.9	0.1102	0.9	0.8533	0.9455
13.8730	7.0	2135	3.3112	0.2862	0.3865	0.3179	0.1296	0.2205	0.3678	0.2994	0.5477	0.5937	0.381	0.4929	0.754	0.7195	0.819	0.5142	0.7446	0.013	0.5889	0.7176	0.7913	0.0006	0.3667	0.0	0.0	0.0333	0.0957	0.4362	0.1206	0.6114	0.2889	0.9	0.1941	0.8875	0.77	0.9455
13.8730	8.0	2440	3.3219	0.299	0.3796	0.3386	0.1301	0.1955	0.39	0.3308	0.5312	0.5921	0.3834	0.4872	0.7547	0.7216	0.8146	0.5243	0.7368	0.0145	0.6	0.7219	0.8008	0.0009	0.3667	0.0	0.0	0.0	0.0998	0.4571	0.0829	0.5904	0.4411	0.9	0.1704	0.875	0.8112	0.9636
13.3734	9.0	2745	3.2589	0.3051	0.3843	0.3451	0.1548	0.2249	0.4004	0.3313	0.5323	0.5906	0.3844	0.5041	0.7497	0.732	0.8218	0.5313	0.7302	0.0122	0.5778	0.7245	0.8038	0.0007	0.3933	0.0286	0.0	0.0	0.1047	0.4466	0.1159	0.6133	0.3143	0.8667	0.3065	0.85	0.8197	0.9545
12.8748	10.0	3050	3.2415	0.327	0.4125	0.3657	0.1554	0.2343	0.4258	0.3385	0.5476	0.5865	0.3892	0.4928	0.758	0.7222	0.8126	0.5262	0.7306	0.0186	0.5556	0.7276	0.8037	0.0007	0.3733	0.0286	0.0	0.0333	0.1234	0.4459	0.1327	0.5989	0.5253	0.8667	0.297	0.825	0.8506	0.9636
12.8748	11.0	3355	3.1340	0.3344	0.4231	0.3757	0.1708	0.2186	0.4306	0.345	0.5461	0.5892	0.3955	0.4908	0.7518	0.7389	0.8212	0.5322	0.7291	0.0193	0.6111	0.7297	0.8037	0.0008	0.3533	0.0	0.0	0.0333	0.1487	0.4828	0.1586	0.5933	0.516	0.8667	0.3194	0.8125	0.8487	0.9636
12.5548	12.0	3660	3.1837	0.3198	0.4027	0.3553	0.1594	0.2213	0.4169	0.3291	0.5514	0.5906	0.3984	0.4985	0.7379	0.738	0.8194	0.5403	0.7313	0.0214	0.5556	0.7197	0.7928	0.0009	0.38	0.05	0.0	0.0667	0.1344	0.4537	0.1632	0.5668	0.3403	0.9	0.3182	0.825	0.8617	0.9455
12.5548	13.0	3965	3.0663	0.3352	0.4354	0.3707	0.1464	0.2284	0.434	0.3557	0.5503	0.5891	0.4076	0.5012	0.7418	0.7178	0.8034	0.5672	0.7244	0.0379	0.5889	0.7233	0.7851	0.0011	0.3533	0.05	0.0	0.0333	0.1503	0.4575	0.1688	0.5857	0.4716	0.9333	0.2871	0.8	0.8977	0.9545
12.2871	14.0	4270	2.7447	0.3379	0.456	0.367	0.16	0.23	0.4328	0.3518	0.5477	0.5937	0.4078	0.4904	0.7623	0.7359	0.8152	0.5891	0.7058	0.077	0.6222	0.7253	0.7932	0.001	0.3867	0.0071	0.0	0.0667	0.1815	0.4799	0.2259	0.569	0.2823	0.9333	0.3158	0.8	0.9204	0.9455
12.0066	15.0	4575	2.6533	0.3644	0.4715	0.411	0.1841	0.2192	0.4815	0.3505	0.56	0.5797	0.4234	0.483	0.7412	0.7416	0.818	0.586	0.6987	0.1121	0.5889	0.73	0.7927	0.0008	0.3133	0.0071	0.0	0.0333	0.1736	0.4567	0.2656	0.5682	0.5393	0.9	0.3274	0.825	0.8967	0.9545
12.0066	16.0	4880	2.6616	0.3588	0.4877	0.3937	0.179	0.239	0.4712	0.3569	0.5524	0.5797	0.3837	0.4934	0.7411	0.7454	0.8261	0.5935	0.7022	0.1274	0.6333	0.7336	0.7903	0.0006	0.32	0.0071	0.0	0.0333	0.1599	0.4444	0.2585	0.5604	0.4957	0.9	0.2904	0.8125	0.9007	0.9273
11.7209	17.0	5185	2.6807	0.3716	0.4777	0.4173	0.1844	0.2469	0.4742	0.3705	0.5492	0.5712	0.3957	0.4843	0.724	0.7256	0.8002	0.5863	0.6995	0.1554	0.6111	0.7309	0.7891	0.001	0.3	0.0214	0.0	0.0333	0.1759	0.453	0.254	0.5435	0.5874	0.8333	0.3428	0.825	0.9001	0.9455
11.7209	18.0	5490	2.6444	0.3709	0.4896	0.4154	0.1788	0.2414	0.4831	0.3555	0.553	0.5859	0.4014	0.4962	0.7498	0.7411	0.8214	0.598	0.702	0.1266	0.6	0.7389	0.7943	0.001	0.3533	0.0214	0.0	0.0333	0.169	0.4519	0.271	0.5606	0.5697	0.9	0.3187	0.8375	0.9172	0.9545
11.6491	19.0	5795	2.6604	0.3668	0.481	0.4116	0.1836	0.2263	0.4786	0.3655	0.5583	0.5922	0.4164	0.4946	0.7551	0.7413	0.8182	0.5941	0.7026	0.1347	0.6333	0.7398	0.797	0.0012	0.3667	0.0143	0.0	0.1333	0.1598	0.4474	0.2691	0.532	0.5764	0.9	0.2918	0.825	0.8936	0.9364
11.4519	20.0	6100	2.6343	0.3565	0.5007	0.38	0.1826	0.2282	0.4782	0.3513	0.5539	0.5922	0.4156	0.5033	0.7353	0.7377	0.8204	0.6022	0.7056	0.1485	0.6111	0.7333	0.7968	0.0007	0.3133	0.0071	0.0001	0.1667	0.1761	0.4653	0.2743	0.5535	0.3778	0.9	0.3122	0.8125	0.9157	0.9545
11.4519	21.0	6405	2.6694	0.3546	0.4562	0.3993	0.1628	0.2369	0.4629	0.3522	0.5588	0.5774	0.4108	0.4819	0.7316	0.7376	0.815	0.5923	0.7013	0.1418	0.6111	0.7357	0.7963	0.0011	0.3333	0.0071	0.0	0.0333	0.1613	0.4474	0.256	0.5008	0.5046	0.9	0.2752	0.8375	0.8496	0.9455
11.2848	22.0	6710	2.6583	0.3456	0.4588	0.3792	0.1871	0.2376	0.4587	0.3347	0.5416	0.5664	0.4101	0.4775	0.7108	0.7393	0.816	0.6033	0.7086	0.1384	0.5889	0.7302	0.787	0.0008	0.28	0.0214	0.0	0.0333	0.1646	0.4414	0.2619	0.4916	0.3609	0.8667	0.2797	0.825	0.8677	0.9364
11.2548	23.0	7015	2.6603	0.3607	0.4761	0.4072	0.1738	0.2398	0.4812	0.3517	0.5465	0.5767	0.4198	0.4737	0.7441	0.7423	0.8172	0.595	0.7013	0.1332	0.6667	0.7295	0.7873	0.0011	0.3067	0.0143	0.0	0.0333	0.1601	0.4347	0.265	0.5098	0.5134	0.9	0.2984	0.8125	0.8903	0.9364
11.2548	24.0	7320	2.6721	0.3583	0.4658	0.3986	0.1656	0.2369	0.4791	0.3593	0.5533	0.5827	0.41	0.491	0.7201	0.7371	0.8122	0.5888	0.6969	0.1378	0.6	0.7378	0.7948	0.0009	0.32	0.0143	0.0001	0.1667	0.1504	0.4198	0.2715	0.4846	0.5046	0.9	0.2803	0.8375	0.8906	0.9455
11.1566	25.0	7625	2.6563	0.3728	0.4796	0.4188	0.1829	0.2411	0.4944	0.3578	0.5475	0.5697	0.4121	0.4792	0.7157	0.7459	0.8204	0.596	0.7	0.1497	0.6	0.7374	0.7923	0.001	0.2867	0.0357	0.0	0.0333	0.1555	0.4455	0.2669	0.4977	0.6144	0.8667	0.313	0.8125	0.8933	0.9455
11.1566	26.0	7930	2.6494	0.3715	0.48	0.4148	0.1917	0.2449	0.4935	0.3587	0.5483	0.5679	0.4082	0.4651	0.7246	0.744	0.82	0.5978	0.7004	0.1646	0.6	0.7395	0.7928	0.001	0.28	0.0214	0.0	0.0333	0.1487	0.4325	0.2854	0.5099	0.6	0.8667	0.2886	0.8125	0.8883	0.9455
11.0968	27.0	8235	2.6704	0.3611	0.4698	0.403	0.1693	0.2442	0.4806	0.3596	0.5513	0.5778	0.4273	0.4816	0.7296	0.7393	0.8184	0.597	0.7042	0.1785	0.6333	0.7386	0.7925	0.0009	0.3067	0.05	0.0	0.0667	0.1487	0.4287	0.2747	0.4992	0.4788	0.8667	0.3095	0.8125	0.8678	0.9545
11.0700	28.0	8540	2.6652	0.3631	0.4658	0.4052	0.1686	0.2453	0.4827	0.3451	0.5341	0.5606	0.4181	0.4731	0.7025	0.7418	0.818	0.5923	0.698	0.1592	0.6111	0.741	0.7975	0.0008	0.2867	0.0143	0.0	0.0333	0.1454	0.4134	0.2787	0.4926	0.6	0.8667	0.2192	0.75	0.8786	0.9455
11.0700	29.0	8845	2.6561	0.3601	0.4632	0.404	0.1807	0.2454	0.4782	0.3591	0.5506	0.5657	0.4224	0.4687	0.7246	0.7426	0.8216	0.5964	0.6985	0.162	0.6111	0.7422	0.7978	0.001	0.2933	0.0143	0.0	0.0	0.1515	0.4216	0.2831	0.5051	0.4715	0.8667	0.3013	0.8125	0.8698	0.9455
11.0346	30.0	9150	2.6627	0.3579	0.4601	0.4002	0.1736	0.2436	0.4745	0.3584	0.543	0.5699	0.4171	0.476	0.7197	0.7402	0.8192	0.5973	0.7	0.1582	0.6222	0.7389	0.7948	0.001	0.2733	0.0214	0.0	0.0667	0.1501	0.4205	0.2746	0.4958	0.4878	0.8667	0.2657	0.8125	0.8808	0.9455

Framework versions

Transformers 5.12.1
Pytorch 2.12.1+cu130
Datasets 5.0.0
Tokenizers 0.22.2

Task: Object Detection
Library: transformers
Type: d_fine
License: apache-2.0
Created: Jun 29, 2026
Updated: Jun 29, 2026