rt_detrv2_finetuned_trashify_box_detector_v1

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: 9.3370
Map: 0.503
Map 50: 0.6852
Map 75: 0.575
Map Small: 0.1001
Map Medium: 0.358
Map Large: 0.523
Mar 1: 0.5764
Mar 10: 0.6893
Mar 100: 0.7409
Mar Small: 0.15
Mar Medium: 0.5585
Mar Large: 0.7722
Map Bin: 0.7562
Mar 100 Bin: 0.8752
Map Hand: 0.6227
Mar 100 Hand: 0.8049
Map Not Bin: 0.1062
Mar 100 Not Bin: 0.6
Map Not Hand: -1.0
Mar 100 Not Hand: -1.0
Map Not Trash: 0.2456
Mar 100 Not Trash: 0.5639
Map Trash: 0.6113
Mar 100 Trash: 0.7681
Map Trash Arm: 0.6757
Mar 100 Trash Arm: 0.8333

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: 0.0001
train_batch_size: 16
eval_batch_size: 16
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: 0.05
num_epochs: 10
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 Bin	Mar 100 Bin	Map Hand	Mar 100 Hand	Map Not Bin	Mar 100 Not Bin	Map Not Hand	Mar 100 Not Hand	Map Not Trash	Mar 100 Not Trash	Map Trash	Mar 100 Trash	Map Trash Arm	Mar 100 Trash Arm
73.4081	1.0	50	21.7482	0.0286	0.0658	0.0214	0.0	0.0028	0.0295	0.0524	0.1063	0.1318	0.0	0.0136	0.1368	0.0272	0.3184	0.1335	0.3676	0.0	0.0	-1.0	-1.0	0.0102	0.0597	0.0007	0.0451	0.0	0.0
35.6688	2.0	100	14.3209	0.224	0.357	0.2415	0.0	0.0516	0.2394	0.2627	0.479	0.5578	0.0	0.1807	0.5867	0.2927	0.727	0.497	0.7912	0.0063	0.4	-1.0	-1.0	0.0964	0.3625	0.4474	0.6327	0.0041	0.4333
28.4572	3.0	150	11.4354	0.3139	0.4683	0.3362	0.0375	0.1049	0.3328	0.3386	0.6328	0.6877	0.15	0.4619	0.7234	0.6222	0.8291	0.5679	0.801	0.0348	0.5214	-1.0	-1.0	0.1421	0.4375	0.5092	0.7372	0.0075	0.8
23.8380	4.0	200	10.1942	0.3711	0.5247	0.4033	0.0286	0.1897	0.3854	0.4682	0.6776	0.7181	0.2	0.5466	0.7448	0.6864	0.873	0.5937	0.8118	0.0403	0.5214	-1.0	-1.0	0.1899	0.5264	0.5746	0.7425	0.1414	0.8333
20.5274	5.0	250	9.5975	0.4582	0.6248	0.5137	0.0273	0.1989	0.4767	0.5608	0.6959	0.7302	0.3	0.4989	0.7556	0.7297	0.8553	0.6249	0.798	0.0591	0.6	-1.0	-1.0	0.2366	0.5292	0.612	0.7655	0.4867	0.8333
17.9319	6.0	300	9.4738	0.4813	0.6518	0.5465	0.0585	0.2543	0.5007	0.5695	0.7099	0.7473	0.2	0.5205	0.7839	0.7482	0.8716	0.5936	0.8167	0.0704	0.6071	-1.0	-1.0	0.2581	0.5972	0.6089	0.7575	0.6084	0.8333
16.2130	7.0	350	9.2680	0.5095	0.6908	0.5815	0.1	0.2732	0.5297	0.5817	0.7012	0.7383	0.1	0.5489	0.7711	0.7521	0.8766	0.623	0.8039	0.0936	0.5786	-1.0	-1.0	0.2701	0.5611	0.6173	0.7761	0.701	0.8333
14.8779	8.0	400	9.2277	0.5111	0.6896	0.5861	0.1	0.324	0.5307	0.5815	0.7151	0.751	0.1	0.5551	0.7802	0.7788	0.8801	0.5978	0.7961	0.0874	0.6071	-1.0	-1.0	0.2399	0.5694	0.6135	0.7867	0.7495	0.8667
13.8448	9.0	450	9.2435	0.4902	0.6684	0.5528	0.05	0.2557	0.5086	0.5801	0.7011	0.7408	0.1	0.5358	0.7709	0.7698	0.8759	0.5967	0.7833	0.1106	0.5857	-1.0	-1.0	0.2448	0.5639	0.6025	0.769	0.6168	0.8667
13.1414	10.0	500	9.3370	0.503	0.6852	0.575	0.1001	0.358	0.523	0.5764	0.6893	0.7409	0.15	0.5585	0.7722	0.7562	0.8752	0.6227	0.8049	0.1062	0.6	-1.0	-1.0	0.2456	0.5639	0.6113	0.7681	0.6757	0.8333

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.6.0
Tokenizers 0.22.2

Loading…

rt_detrv2_finetuned_trashify_box_detector_v1

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: 9.3370

Map: 0.503

Map 50: 0.6852

Map 75: 0.575

Map Small: 0.1001

Map Medium: 0.358

Map Large: 0.523

Mar 1: 0.5764

Mar 10: 0.6893

Mar 100: 0.7409

Mar Small: 0.15

Mar Medium: 0.5585

Mar Large: 0.7722

Map Bin: 0.7562

Mar 100 Bin: 0.8752

Map Hand: 0.6227

Mar 100 Hand: 0.8049

Map Not Bin: 0.1062

Mar 100 Not Bin: 0.6

Map Not Hand: -1.0

Mar 100 Not Hand: -1.0

Map Not Trash: 0.2456

Mar 100 Not Trash: 0.5639

Map Trash: 0.6113

Mar 100 Trash: 0.7681

Map Trash Arm: 0.6757

Mar 100 Trash Arm: 0.8333

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0001

train_batch_size: 16

eval_batch_size: 16

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: 0.05

num_epochs: 10

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 Bin	Mar 100 Bin	Map Hand	Mar 100 Hand	Map Not Bin	Mar 100 Not Bin	Map Not Hand	Mar 100 Not Hand	Map Not Trash	Mar 100 Not Trash	Map Trash	Mar 100 Trash	Map Trash Arm	Mar 100 Trash Arm
73.4081	1.0	50	21.7482	0.0286	0.0658	0.0214	0.0	0.0028	0.0295	0.0524	0.1063	0.1318	0.0	0.0136	0.1368	0.0272	0.3184	0.1335	0.3676	0.0	0.0	-1.0	-1.0	0.0102	0.0597	0.0007	0.0451	0.0	0.0
35.6688	2.0	100	14.3209	0.224	0.357	0.2415	0.0	0.0516	0.2394	0.2627	0.479	0.5578	0.0	0.1807	0.5867	0.2927	0.727	0.497	0.7912	0.0063	0.4	-1.0	-1.0	0.0964	0.3625	0.4474	0.6327	0.0041	0.4333
28.4572	3.0	150	11.4354	0.3139	0.4683	0.3362	0.0375	0.1049	0.3328	0.3386	0.6328	0.6877	0.15	0.4619	0.7234	0.6222	0.8291	0.5679	0.801	0.0348	0.5214	-1.0	-1.0	0.1421	0.4375	0.5092	0.7372	0.0075	0.8
23.8380	4.0	200	10.1942	0.3711	0.5247	0.4033	0.0286	0.1897	0.3854	0.4682	0.6776	0.7181	0.2	0.5466	0.7448	0.6864	0.873	0.5937	0.8118	0.0403	0.5214	-1.0	-1.0	0.1899	0.5264	0.5746	0.7425	0.1414	0.8333
20.5274	5.0	250	9.5975	0.4582	0.6248	0.5137	0.0273	0.1989	0.4767	0.5608	0.6959	0.7302	0.3	0.4989	0.7556	0.7297	0.8553	0.6249	0.798	0.0591	0.6	-1.0	-1.0	0.2366	0.5292	0.612	0.7655	0.4867	0.8333
17.9319	6.0	300	9.4738	0.4813	0.6518	0.5465	0.0585	0.2543	0.5007	0.5695	0.7099	0.7473	0.2	0.5205	0.7839	0.7482	0.8716	0.5936	0.8167	0.0704	0.6071	-1.0	-1.0	0.2581	0.5972	0.6089	0.7575	0.6084	0.8333
16.2130	7.0	350	9.2680	0.5095	0.6908	0.5815	0.1	0.2732	0.5297	0.5817	0.7012	0.7383	0.1	0.5489	0.7711	0.7521	0.8766	0.623	0.8039	0.0936	0.5786	-1.0	-1.0	0.2701	0.5611	0.6173	0.7761	0.701	0.8333
14.8779	8.0	400	9.2277	0.5111	0.6896	0.5861	0.1	0.324	0.5307	0.5815	0.7151	0.751	0.1	0.5551	0.7802	0.7788	0.8801	0.5978	0.7961	0.0874	0.6071	-1.0	-1.0	0.2399	0.5694	0.6135	0.7867	0.7495	0.8667
13.8448	9.0	450	9.2435	0.4902	0.6684	0.5528	0.05	0.2557	0.5086	0.5801	0.7011	0.7408	0.1	0.5358	0.7709	0.7698	0.8759	0.5967	0.7833	0.1106	0.5857	-1.0	-1.0	0.2448	0.5639	0.6025	0.769	0.6168	0.8667
13.1414	10.0	500	9.3370	0.503	0.6852	0.575	0.1001	0.358	0.523	0.5764	0.6893	0.7409	0.15	0.5585	0.7722	0.7562	0.8752	0.6227	0.8049	0.1062	0.6	-1.0	-1.0	0.2456	0.5639	0.6113	0.7681	0.6757	0.8333

Framework versions

Transformers 5.0.0

Pytorch 2.10.0+cu128

Datasets 4.6.0

Tokenizers 0.22.2

NickNick414/rt_detrv2_finetuned_trashify_box_detector_v1

rt_detrv2_finetuned_trashify_box_detector_v1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

rt_detrv2_finetuned_trashify_box_detector_v1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions