Hugging Face's logo

mfly-auton
/
suas-2025-rtdetr-finetuned-b8-lr3e-5

Object Detection
Transformers
Safetensors
rt_detr
vision
Generated from Trainer
Inference Endpoints
Model card Files Community
suas-2025-rtdetr-finetuned-b8-lr3e-5 / README.md
emvisee's picture
emvisee
End of training
897e33e verified
preview code
|
raw
history blame contribute delete
10.8 kB
metadata
library_name: transformers
license: apache-2.0
base_model: PekingU/rtdetr_r50vd
tags:
  - object-detection
  - vision
  - generated_from_trainer
model-index:
  - name: suas-2025-rtdetr-finetuned-b8-lr3e-5
    results: []

suas-2025-rtdetr-finetuned-b8-lr3e-5

This model is a fine-tuned version of PekingU/rtdetr_r50vd on the mfly-auton/suas-2025-synthetic-data dataset. It achieves the following results on the evaluation set:

  • Loss: 6.4174
  • Map: 0.6051
  • Map 50: 0.7291
  • Map 75: 0.6597
  • Map Small: 0.5252
  • Map Medium: 0.6131
  • Map Large: 0.6723
  • Mar 1: 0.6006
  • Mar 10: 0.7415
  • Mar 100: 0.7959
  • Mar Small: 0.5831
  • Mar Medium: 0.8299
  • Mar Large: 0.948
  • Map Baseball-bat: 0.5841
  • Mar 100 Baseball-bat: 0.7219
  • Map Basketball: 0.4357
  • Mar 100 Basketball: 0.4955
  • Map Car: -1.0
  • Mar 100 Car: -1.0
  • Map Football: 0.4363
  • Mar 100 Football: 0.5014
  • Map Human: 0.9189
  • Mar 100 Human: 0.9512
  • Map Luggage: 0.5995
  • Mar 100 Luggage: 0.8773
  • Map Mattress: 0.7881
  • Mar 100 Mattress: 0.9906
  • Map Motorcycle: 0.9485
  • Mar 100 Motorcycle: 0.9752
  • Map Skis: 0.9582
  • Mar 100 Skis: 0.9782
  • Map Snowboard: 0.0257
  • Mar 100 Snowboard: 0.777
  • Map Soccer-ball: 0.6432
  • Mar 100 Soccer-ball: 0.7154
  • Map Stop-sign: 0.5994
  • Mar 100 Stop-sign: 0.9274
  • Map Tennis-racket: 0.0019
  • Mar 100 Tennis-racket: 0.5736
  • Map Umbrella: 0.8665
  • Mar 100 Umbrella: 0.9243
  • Map Volleyball: 0.6647
  • Mar 100 Volleyball: 0.7338

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: 3e-05
  • train_batch_size: 8
  • eval_batch_size: 32
  • seed: 1337
  • 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
  • num_epochs: 10.0
  • 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 Baseball-bat Mar 100 Baseball-bat Map Basketball Mar 100 Basketball Map Car Mar 100 Car Map Football Mar 100 Football Map Human Mar 100 Human Map Luggage Mar 100 Luggage Map Mattress Mar 100 Mattress Map Motorcycle Mar 100 Motorcycle Map Skis Mar 100 Skis Map Snowboard Mar 100 Snowboard Map Soccer-ball Mar 100 Soccer-ball Map Stop-sign Mar 100 Stop-sign Map Tennis-racket Mar 100 Tennis-racket Map Umbrella Mar 100 Umbrella Map Volleyball Mar 100 Volleyball
17.7551 1.0 875 5.2510 0.6942 0.7838 0.7782 0.7254 0.6405 0.7363 0.7413 0.8942 0.9051 0.8089 0.8996 0.9496 0.6964 0.8173 0.7915 0.8415 -1.0 -1.0 0.7303 0.8192 0.8746 0.9281 0.6526 0.9372 0.9142 0.9711 0.8867 0.9311 0.8376 0.9594 0.3429 0.9742 0.7454 0.8686 0.581 0.9542 0.3918 0.944 0.4713 0.8414 0.8027 0.8846
6.6319 2.0 1750 5.6671 0.6915 0.7796 0.7715 0.6499 0.6468 0.8104 0.7323 0.879 0.892 0.726 0.9168 0.9841 0.6895 0.7787 0.5032 0.6076 -1.0 -1.0 0.7195 0.7926 0.9186 0.9593 0.6184 0.9288 0.895 0.9984 0.901 0.9643 0.9222 0.9634 0.2897 0.9699 0.7717 0.8578 0.6219 0.9584 0.2013 0.8995 0.8724 0.9458 0.7562 0.864
6.3092 3.0 2625 5.3786 0.6861 0.7797 0.7703 0.6794 0.6505 0.8039 0.7008 0.8593 0.8895 0.738 0.9294 0.9851 0.7854 0.863 0.5392 0.5953 -1.0 -1.0 0.6451 0.7011 0.9165 0.9458 0.6485 0.9354 0.9011 0.9986 0.9194 0.9592 0.9354 0.9827 0.147 0.9815 0.8089 0.8686 0.7102 0.9732 0.0527 0.8694 0.8651 0.9418 0.7312 0.838
5.8652 4.0 3500 5.8538 0.6465 0.749 0.7306 0.6014 0.6486 0.6801 0.6644 0.8092 0.8383 0.6448 0.8784 0.9506 0.6314 0.7432 0.4033 0.446 -1.0 -1.0 0.6013 0.6536 0.9291 0.9609 0.729 0.9151 0.64 0.9942 0.9361 0.9592 0.8382 0.9396 0.1472 0.893 0.7465 0.8059 0.803 0.952 0.0352 0.7725 0.8569 0.892 0.7539 0.8091
5.4748 5.0 4375 5.8922 0.6416 0.7467 0.7203 0.5884 0.6359 0.7155 0.6543 0.8104 0.8498 0.6337 0.895 0.965 0.6726 0.7857 0.3624 0.4249 -1.0 -1.0 0.5819 0.641 0.9266 0.9518 0.6374 0.9073 0.7673 0.9959 0.9372 0.9692 0.9629 0.9861 0.0824 0.9316 0.7745 0.8198 0.7208 0.9508 0.015 0.8404 0.7927 0.8863 0.7489 0.8061
5.352 6.0 5250 6.3142 0.6409 0.741 0.7178 0.6313 0.5958 0.6861 0.6395 0.7987 0.8411 0.6741 0.8595 0.9544 0.7148 0.8051 0.4792 0.5386 -1.0 -1.0 0.5879 0.6392 0.9101 0.9478 0.5811 0.8811 0.8213 0.9987 0.9139 0.96 0.9479 0.9767 0.052 0.9105 0.7854 0.8321 0.6801 0.9332 0.0209 0.6995 0.7664 0.8798 0.7112 0.7733
5.1405 7.0 6125 6.3704 0.6464 0.7579 0.7185 0.5862 0.6209 0.6971 0.6346 0.7737 0.8247 0.6346 0.865 0.9372 0.6828 0.7855 0.5464 0.5882 -1.0 -1.0 0.5058 0.5697 0.9172 0.9526 0.6302 0.9021 0.8037 0.9908 0.9339 0.9643 0.9615 0.9767 0.0249 0.6942 0.7298 0.7859 0.7449 0.9567 0.0025 0.713 0.8942 0.9397 0.6718 0.7272
4.9059 8.0 7000 6.4879 0.6238 0.7487 0.6799 0.5369 0.6248 0.6895 0.6205 0.7655 0.8153 0.5983 0.8634 0.9561 0.6155 0.7369 0.4736 0.5339 -1.0 -1.0 0.4614 0.5228 0.9159 0.9516 0.6436 0.8783 0.867 0.9951 0.9482 0.9714 0.9574 0.9767 0.0317 0.806 0.6742 0.7409 0.6387 0.9321 0.0037 0.7368 0.8625 0.9115 0.6398 0.7199
4.8446 9.0 7875 6.5165 0.6064 0.734 0.6583 0.5305 0.6055 0.6766 0.6057 0.7479 0.8021 0.5908 0.8441 0.9489 0.598 0.7249 0.4926 0.5547 -1.0 -1.0 0.4304 0.4976 0.9153 0.9478 0.5701 0.8526 0.7636 0.9867 0.9459 0.9684 0.9583 0.9762 0.0271 0.7862 0.6457 0.7131 0.6268 0.9338 0.0014 0.6518 0.8662 0.914 0.6484 0.7213
4.7656 10.0 8750 6.4174 0.6051 0.7291 0.6597 0.5252 0.6131 0.6723 0.6006 0.7415 0.7959 0.5831 0.8299 0.948 0.5841 0.7219 0.4357 0.4955 -1.0 -1.0 0.4363 0.5014 0.9189 0.9512 0.5995 0.8773 0.7881 0.9906 0.9485 0.9752 0.9582 0.9782 0.0257 0.777 0.6432 0.7154 0.5994 0.9274 0.0019 0.5736 0.8665 0.9243 0.6647 0.7338

Framework versions

  • Transformers 4.47.0
  • Pytorch 2.5.1+cu124
  • Datasets 3.1.0
  • Tokenizers 0.21.0