README.md

metadata

library_name: transformers
license: apache-2.0
base_model: PekingU/rtdetr_v2_r50vd
tags:
  - generated_from_trainer
model-index:
  - name: learn_hf-rt-detrv2-finetuned-on-trashify-dataset-video
    results: []

learn_hf-rt-detrv2-finetuned-on-trashify-dataset-video

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: 10.1612
• Map: 0.4478
• Map 50: 0.5864
• Map 75: 0.5096
• Map Small: 0.0
• Map Medium: 0.2891
• Map Large: 0.4581
• Mar 1: 0.5121
• Mar 10: 0.7092
• Mar 100: 0.7613
• Mar Small: 0.0
• Mar Medium: 0.5975
• Mar Large: 0.7815
• Map Bin: 0.746
• Mar Bin: 0.9187
• Map Hand: 0.5961
• Mar Hand: 0.8136
• Map Not Bin: 0.0561
• Mar Not Bin: 0.5727
• Map Not Hand: 0.0185
• Mar Not Hand: 0.6333
• Map Not Trash: 0.2151
• Mar Not Trash: 0.6222
• Map Trash: 0.6585
• Mar Trash: 0.8397
• Map Trash Arm: 0.8444
• Mar Trash Arm: 0.9286

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 with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 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 Bin	Map Hand	Mar Hand	Map Not Bin	Mar Not Bin	Map Not Hand	Mar Not Hand	Map Not Trash	Mar Not Trash	Map Trash	Mar Trash	Map Trash Arm	Mar Trash Arm
90.2652	1.0	50	22.7169	0.2343	0.342	0.2447	0.0	0.0133	0.2432	0.3225	0.502	0.5748	0.0	0.0977	0.628	0.5467	0.8326	0.4655	0.65	0.007	0.45	-1.0	-1.0	0.0117	0.3264	0.3706	0.623	0.0044	0.5667
26.5507	2.0	100	13.5049	0.4103	0.5773	0.4496	0.02	0.1152	0.4293	0.4729	0.6547	0.7171	0.1	0.3898	0.7515	0.6658	0.8965	0.5505	0.7667	0.008	0.5071	-1.0	-1.0	0.1584	0.5111	0.6138	0.7876	0.4657	0.8333
19.0339	3.0	150	11.7506	0.4485	0.6168	0.528	0.0714	0.2235	0.4651	0.5086	0.7004	0.7429	0.25	0.5403	0.7789	0.6937	0.8894	0.596	0.7931	0.0112	0.55	-1.0	-1.0	0.1558	0.5931	0.6348	0.7982	0.5993	0.8333
16.4727	4.0	200	11.0906	0.5126	0.6896	0.5739	0.0	0.2155	0.5353	0.5473	0.7081	0.7555	0.0	0.4216	0.798	0.7116	0.8851	0.6232	0.7922	0.0478	0.55	-1.0	-1.0	0.2404	0.5819	0.6315	0.7903	0.8211	0.9333
15.0827	5.0	250	10.7144	0.4955	0.6776	0.5712	0.125	0.225	0.518	0.5428	0.695	0.7541	0.25	0.4699	0.7957	0.7618	0.9113	0.5424	0.7735	0.0344	0.5357	-1.0	-1.0	0.248	0.5986	0.6267	0.8053	0.7599	0.9
13.9634	6.0	300	10.6771	0.5377	0.7197	0.5907	0.2	0.3086	0.567	0.571	0.737	0.7894	0.2	0.5352	0.8348	0.7489	0.9121	0.5976	0.7951	0.1615	0.7214	-1.0	-1.0	0.2387	0.6	0.6643	0.808	0.8148	0.9
13.0714	7.0	350	10.4076	0.5525	0.7296	0.6065	0.2	0.1863	0.5876	0.5696	0.7384	0.781	0.2	0.3295	0.8492	0.7707	0.9078	0.629	0.8127	0.1764	0.5929	-1.0	-1.0	0.2363	0.5889	0.657	0.8168	0.8456	0.9667
12.405	8.0	400	10.2652	0.5346	0.7063	0.6085	0.3	0.2124	0.5697	0.55	0.7165	0.7691	0.3	0.3716	0.8306	0.7666	0.9028	0.5832	0.8118	0.1874	0.5786	-1.0	-1.0	0.2255	0.6153	0.644	0.8062	0.8007	0.9
11.7512	9.0	450	10.0407	0.5506	0.7358	0.6293	0.3	0.2311	0.5877	0.5614	0.7456	0.7755	0.3	0.4398	0.8337	0.7602	0.9142	0.6356	0.8029	0.2287	0.5857	-1.0	-1.0	0.2414	0.6306	0.6674	0.8195	0.7705	0.9
11.3543	10.0	500	10.0798	0.5401	0.7284	0.6267	0.3	0.2092	0.5756	0.5473	0.7485	0.7801	0.3	0.3767	0.8405	0.7629	0.905	0.6323	0.8059	0.2095	0.5857	-1.0	-1.0	0.2192	0.6278	0.6597	0.823	0.7569	0.9333

Framework versions

• Transformers 4.53.1
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.2