rtdetr-v2-r50-lpph-finetune-2

ranm26/rtdetr-v2-r50-lpph-finetune-2

rtdetr-v2-r50-lpph-finetune-2 is a object detection model from ranm26. 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

112

14% of all-time downloads

All-time downloads

800

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

Task: Object Detection
Library: transformers
License: apache-2.0
Model type: rt_detr_v2
Access: Public
Created: Nov 15, 2025
Updated: Nov 23, 2025
SHA: 4ca3df20

Base models

PekingU/rtdetr_v2_r50vd

Model card

From the Hugging Face model README

rtdetr-v2-r50-lpph-finetune-2

This model is a fine-tuned version of PekingU/rtdetr_v2_r50vd on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 6.8109
Map: 0.6252
Mar 100: 0.7345
Map 50: 0.9661
Map 75: 0.7201

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: 2e-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
lr_scheduler_warmup_steps: 300
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Mar 100	Map 50	Map 75
No log	1.0	142	375.5752	0.0	0.0085	0.0001	0.0
No log	2.0	284	180.1939	0.0442	0.5191	0.0797	0.0475
No log	3.0	426	58.7968	0.5836	0.6979	0.9564	0.6474
300.3878	4.0	568	17.9708	0.5889	0.7021	0.9645	0.6757
300.3878	5.0	710	9.5945	0.5882	0.683	0.9596	0.7067
300.3878	6.0	852	7.5412	0.5918	0.7121	0.9715	0.6658
300.3878	7.0	994	7.0515	0.6004	0.7191	0.9803	0.7103
17.3345	8.0	1136	6.7723	0.6205	0.7369	0.9764	0.7385
17.3345	9.0	1278	6.5420	0.6245	0.7376	0.9784	0.7714
17.3345	10.0	1420	6.6140	0.625	0.7376	0.9723	0.7721
11.8344	11.0	1562	6.6079	0.6075	0.7312	0.9718	0.7216
11.8344	12.0	1704	6.5404	0.6062	0.7106	0.9688	0.7289
11.8344	13.0	1846	6.5771	0.6091	0.7128	0.9744	0.7282
11.8344	14.0	1988	6.3854	0.619	0.7397	0.9706	0.7486
10.9366	15.0	2130	6.4225	0.6188	0.739	0.973	0.7735
10.9366	16.0	2272	6.5708	0.6141	0.7227	0.9714	0.7274
10.9366	17.0	2414	6.7743	0.6099	0.7298	0.9572	0.7796
10.2994	18.0	2556	6.5782	0.605	0.7241	0.9681	0.7289
10.2994	19.0	2698	6.7827	0.6167	0.7199	0.9675	0.7293
10.2994	20.0	2840	6.5982	0.6143	0.7177	0.9749	0.7197
10.2994	21.0	2982	6.7339	0.6064	0.7206	0.9713	0.73
9.6559	22.0	3124	6.7611	0.6201	0.7078	0.9727	0.7747
9.6559	23.0	3266	6.5581	0.6222	0.7284	0.9754	0.7924
9.6559	24.0	3408	6.6518	0.6226	0.717	0.9743	0.791
9.1057	25.0	3550	6.5741	0.6202	0.7227	0.9769	0.7856
9.1057	26.0	3692	6.6062	0.6211	0.7234	0.979	0.7408
9.1057	27.0	3834	6.6890	0.6213	0.7319	0.9768	0.7764
9.1057	28.0	3976	6.6388	0.6123	0.7121	0.9793	0.77
8.591	29.0	4118	6.6831	0.6128	0.7206	0.9787	0.7518
8.591	30.0	4260	6.6156	0.6192	0.7255	0.9746	0.7833
8.591	31.0	4402	6.8166	0.6132	0.7142	0.9771	0.752
8.2786	32.0	4544	6.6713	0.6135	0.727	0.9777	0.767
8.2786	33.0	4686	6.7634	0.6202	0.7213	0.9792	0.789
8.2786	34.0	4828	6.7925	0.6152	0.7227	0.9759	0.7832
8.2786	35.0	4970	6.8108	0.612	0.7227	0.9769	0.7724
7.8299	36.0	5112	6.8419	0.6113	0.7227	0.9768	0.7742
7.8299	37.0	5254	6.8886	0.6094	0.7248	0.9778	0.7504
7.8299	38.0	5396	6.8877	0.6101	0.717	0.9785	0.775
7.5879	39.0	5538	6.9249	0.6084	0.7177	0.9763	0.7666
7.5879	40.0	5680	6.9350	0.606	0.7156	0.9767	0.7544

Framework versions

Transformers 4.57.1
Pytorch 2.9.1+cu128
Datasets 4.4.1
Tokenizers 0.22.1

Task: Object Detection
Library: transformers
Type: rt_detr_v2
License: apache-2.0
Created: Nov 15, 2025
Updated: Nov 23, 2025