rtdetr-r50-fruits2-finetune

This model is a fine-tuned version of PekingU/rtdetr_r50vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 12.5005
• Map: 0.4374
• Map 50: 0.5647
• Map 75: 0.4601
• Map Small: 0.1357
• Map Medium: 0.3572
• Map Large: 0.617
• Mar 1: 0.2516
• Mar 10: 0.5739
• Mar 100: 0.7003
• Mar Small: 0.3578
• Mar Medium: 0.6855
• Mar Large: 0.8642
• Map Apple: 0.4334
• Mar 100 Apple: 0.6864
• Map Banana: 0.4636
• Mar 100 Banana: 0.7153
• Map Grapes: 0.4037
• Mar 100 Grapes: 0.6071
• Map Orange: 0.305
• Mar 100 Orange: 0.6301
• Map Pineapple: 0.5309
• Mar 100 Pineapple: 0.7449
• Map Watermelon: 0.4881
• Mar 100 Watermelon: 0.8182

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 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: 1

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 Apple	Mar 100 Apple	Map Banana	Mar 100 Banana	Map Grapes	Mar 100 Grapes	Map Orange	Mar 100 Orange	Map Pineapple	Mar 100 Pineapple	Map Watermelon	Mar 100 Watermelon
35.0679	1.0	750	12.5005	0.4374	0.5647	0.4601	0.1357	0.3572	0.617	0.2516	0.5739	0.7003	0.3578	0.6855	0.8642	0.4334	0.6864	0.4636	0.7153	0.4037	0.6071	0.305	0.6301	0.5309	0.7449	0.4881	0.8182

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1