rtdetr3

b09501048/rtdetr3

rtdetr3 is a object detection model from b09501048. 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

3% of all-time downloads

All-time downloads

404

Public

Parameters

42.9M

858 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors172 MB · 100%

At a glance

Task: Object Detection
Library: transformers
License: apache-2.0
Model type: rt_detr
Access: Public
Created: Oct 13, 2024
Updated: Oct 13, 2024
SHA: 3439e54b

Base models

b09501048/rtdetr2

Model card

From the Hugging Face model README

rtdetr3

This model is a fine-tuned version of b09501048/rtdetr2 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 12.0504
Map: 0.2831
Map 50: 0.4512
Map 75: 0.2953
Map Small: 0.1426
Map Medium: 0.3859
Map Large: 0.469
Mar 1: 0.2299
Mar 10: 0.441
Mar 100: 0.4546
Mar Small: 0.2539
Mar Medium: 0.5608
Mar Large: 0.6712
Map Person: 0.716
Mar 100 Person: 0.8309
Map Ear: 0.357
Mar 100 Ear: 0.459
Map Earmuffs: 0.2616
Mar 100 Earmuffs: 0.4676
Map Face: 0.4488
Mar 100 Face: 0.6417
Map Face-guard: 0.2019
Mar 100 Face-guard: 0.5125
Map Face-mask-medical: 0.1874
Mar 100 Face-mask-medical: 0.2842
Map Foot: 0.0467
Mar 100 Foot: 0.2941
Map Tools: 0.0755
Mar 100 Tools: 0.2582
Map Glasses: 0.2358
Mar 100 Glasses: 0.3972
Map Gloves: 0.2782
Mar 100 Gloves: 0.4345
Map Helmet: 0.2651
Mar 100 Helmet: 0.3929
Map Hands: 0.5228
Mar 100 Hands: 0.6237
Map Head: 0.571
Mar 100 Head: 0.6738
Map Medical-suit: 0.1583
Mar 100 Medical-suit: 0.45
Map Shoes: 0.3526
Mar 100 Shoes: 0.4729
Map Safety-suit: 0.0216
Mar 100 Safety-suit: 0.35
Map Safety-vest: 0.1126
Mar 100 Safety-vest: 0.1855

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 adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 300
num_epochs: 5
mixed_precision_training: Native AMP

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 Ear	Mar 100 Ear	Map Earmuffs	Mar 100 Earmuffs	Map Face	Mar 100 Face	Map Face-guard	Mar 100 Face-guard	Map Face-mask-medical	Mar 100 Face-mask-medical	Map Foot	Mar 100 Foot	Map Tools	Mar 100 Tools	Map Glasses	Mar 100 Glasses	Map Gloves	Mar 100 Gloves	Map Helmet	Mar 100 Helmet	Map Hands	Mar 100 Hands	Map Head	Mar 100 Head	Map Medical-suit	Mar 100 Medical-suit	Map Shoes	Mar 100 Shoes	Map Safety-suit	Mar 100 Safety-suit	Map Safety-vest	Mar 100 Safety-vest
No log	1.0	459	12.7473	0.2463	0.3958	0.2544	0.1222	0.3462	0.3923	0.2108	0.3909	0.404	0.2092	0.4925	0.6225	0.6449	0.827	0.3065	0.4	0.1965	0.4378	0.4368	0.603	0.1483	0.45	0.1711	0.2474	0.0568	0.2373	0.0683	0.2296	0.2028	0.3113	0.2313	0.3721	0.2246	0.35	0.5063	0.6106	0.5744	0.672	0.0043	0.2778	0.3112	0.4452	0.0066	0.235	0.0971	0.1613
14.0301	2.0	918	11.8304	0.2764	0.4391	0.283	0.141	0.3804	0.4657	0.2313	0.4441	0.4637	0.244	0.5718	0.7022	0.6841	0.8373	0.3857	0.4744	0.2263	0.4865	0.4943	0.6643	0.2173	0.575	0.1887	0.3298	0.0288	0.2686	0.0586	0.2647	0.2346	0.4232	0.3175	0.4628	0.2462	0.4098	0.5093	0.6097	0.6284	0.7096	0.0157	0.3778	0.3542	0.4786	0.0162	0.34	0.0935	0.171
13.9816	3.0	1377	12.0861	0.2814	0.4484	0.2909	0.141	0.3911	0.4422	0.2314	0.4479	0.4649	0.2391	0.5663	0.6917	0.7053	0.8295	0.3844	0.4867	0.2688	0.4703	0.426	0.6411	0.1494	0.475	0.1747	0.3035	0.0632	0.3275	0.0788	0.2598	0.2417	0.409	0.2273	0.3774	0.296	0.417	0.5231	0.6358	0.6	0.6995	0.1266	0.5167	0.3678	0.4991	0.0182	0.31	0.1321	0.2452
13.4376	4.0	1836	12.0504	0.2831	0.4512	0.2953	0.1426	0.3859	0.469	0.2299	0.441	0.4546	0.2539	0.5608	0.6712	0.716	0.8309	0.357	0.459	0.2616	0.4676	0.4488	0.6417	0.2019	0.5125	0.1874	0.2842	0.0467	0.2941	0.0755	0.2582	0.2358	0.3972	0.2782	0.4345	0.2651	0.3929	0.5228	0.6237	0.571	0.6738	0.1583	0.45	0.3526	0.4729	0.0216	0.35	0.1126	0.1855
13.0503	5.0	2295	12.2470	0.2781	0.4505	0.2859	0.1352	0.3917	0.4434	0.2417	0.4386	0.4518	0.2304	0.5572	0.6509	0.7016	0.8247	0.3341	0.4344	0.2608	0.4919	0.4105	0.6234	0.1222	0.4375	0.1875	0.2825	0.0837	0.302	0.0757	0.2683	0.2379	0.3938	0.2675	0.4119	0.295	0.4045	0.5211	0.6243	0.5795	0.6805	0.131	0.4833	0.3578	0.4833	0.0333	0.345	0.1284	0.1887

Framework versions

Transformers 4.46.0.dev0
Pytorch 2.4.1+cu121
Datasets 3.0.1
Tokenizers 0.20.1

Task: Object Detection
Library: transformers
Type: rt_detr
License: apache-2.0
Created: Oct 13, 2024
Updated: Oct 13, 2024