CoolFace
Modelpublic

paulbauriegel/rtdetr_v2_r101vd-rocks-finetune

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face
0likes14downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

rtdetrv2r101vd-rocks-finetune

This model is a fine-tuned version of PekingU/rtdetr_v2_r101vd on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 12.1315
  • —Map: 0.6911
  • —Map 50: 0.9158
  • —Map 75: 0.9158
  • —Map Small: 0.6597
  • —Map Medium: 0.7
  • —Map Large: -1.0
  • —Mar 1: 0.5
  • —Mar 10: 0.8333
  • —Mar 100: 0.8333
  • —Mar Small: 0.9
  • —Mar Medium: 0.7
  • —Mar Large: -1.0
  • —Map Stone: 0.6911
  • —Mar 100 Stone: 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: 5e-05
  • —trainbatchsize: 8
  • —evalbatchsize: 8
  • —seed: 42
  • —optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 300
  • —num_epochs: 45

Training results

Training LossEpochStepValidation LossMapMap 50Map 75Map SmallMap MediumMap LargeMar 1Mar 10Mar 100Mar SmallMar MediumMar LargeMap StoneMar 100 Stone
No log1.013043.37170.02760.04230.03010.03360.03130.03330.04160.41880.5650.56090.55640.57930.02760.565
No log2.026019.53030.1990.27470.23350.51230.2280.02690.19210.5350.61190.69360.5950.53260.1990.6119
No log3.039014.18440.13910.20290.15480.13550.18130.13630.24750.57290.62870.65180.62280.60760.13910.6287
76.1784.052014.11970.24430.33650.27040.34430.28980.17330.2350.57920.62640.66360.62080.5880.24430.6264
76.1785.065014.18670.14130.19630.16230.17730.17020.15080.13890.55310.64980.69090.64950.60110.14130.6498
76.1786.078014.07050.2170.30.24590.34370.23120.12590.19340.55480.64880.69730.63860.60220.2170.6488
76.1787.091014.43450.10720.1640.11460.11830.14080.11430.19170.54720.6650.73730.65840.58590.10720.665
13.18618.0104014.16360.0930.13540.09940.13550.10110.14150.17950.52240.64390.69730.63860.58590.0930.6439
13.18619.0117014.21290.24640.33660.27990.42910.24590.08260.19310.56770.66010.71820.63470.61850.24640.6601
13.186110.0130014.28450.32640.45590.36180.5180.30470.0940.24920.60330.65640.70450.64360.6130.32640.6564
13.186111.0143013.87620.26870.36880.31450.4330.28360.10720.22510.56270.66830.720.66140.61410.26870.6683
11.763912.0156013.93860.23060.31740.2730.33770.31470.130.2320.55310.65020.69820.63960.60430.23060.6502
11.763913.0169014.24540.17760.26630.19570.32210.16090.04570.18090.52110.65120.68090.65540.61090.17760.6512
11.763914.0182014.39470.21220.29650.23810.34110.25660.09550.22670.55710.65310.70910.62670.61520.21220.6531
11.763915.0195014.40630.27110.37090.31060.41070.3070.10420.2320.57230.64820.70270.62480.60870.27110.6482
11.038716.0208014.75600.24640.34460.28480.40270.26420.11130.20730.55910.64520.69550.63960.59130.24640.6452
11.038717.0221014.36570.23770.32130.27790.40690.21850.06910.16830.5620.67760.72550.66930.62930.23770.6776
11.038718.0234014.73380.12820.17870.14360.22570.09180.07710.15710.50360.65840.70910.64650.61090.12820.6584
11.038719.0247013.88300.22780.32850.25490.38160.24290.08080.20690.57260.6650.71360.64360.63040.22780.665
10.307120.0260013.96960.23590.32830.26780.34420.25890.20310.24520.59310.67260.73820.64750.62170.23590.6726
10.307121.0273014.12700.21960.30310.25250.35770.21660.12950.20790.5320.65910.73090.62670.60870.21960.6591
10.307122.0286014.38960.22090.32640.25740.3960.16520.08870.17690.51820.64590.70550.63470.5870.22090.6459
10.307123.0299014.46890.22340.32490.24360.33880.26670.0710.20330.54260.66240.70360.64460.63260.22340.6624
9.657424.0312014.66900.21650.29910.25330.35430.2250.09440.21820.54820.66370.72450.64550.61090.21650.6637
9.657425.0325014.72140.25110.34170.28270.40540.29580.08780.20660.54460.64690.70820.5990.62610.25110.6469
9.657426.0338014.48130.23020.32650.26560.38930.21840.10560.18910.51620.65280.71550.61580.61850.23020.6528
9.043327.0351014.54060.24070.32450.28030.37730.2680.12870.20760.53470.64820.71450.62770.59130.24070.6482
9.043328.0364014.45530.20260.28860.23810.29770.23680.08880.21390.53960.65020.70.64060.60110.20260.6502
9.043329.0377014.79260.20450.29020.22950.3220.21660.09710.21090.51720.63860.70.61090.59570.20450.6386
9.043330.0390014.99660.22890.32170.26190.38520.2430.06610.21680.50830.65480.71270.65150.58910.22890.6548
8.520631.0403014.84640.22380.32360.24940.39190.19910.06350.19080.51320.64190.71090.62670.57610.22380.6419
8.520632.0416014.92700.24270.34870.27980.3950.26520.08470.21620.51820.6320.71090.6020.57070.24270.632
8.520633.0429014.68490.23580.32820.27630.39720.22810.08930.20460.53170.65210.72820.62280.59350.23580.6521
8.520634.0442014.73760.26750.38310.30250.45770.2380.0870.22510.54980.64950.71730.61680.60430.26750.6495
7.992935.0455014.71600.24740.34830.280.42880.22260.08570.19830.5340.64550.72180.61780.58480.24740.6455
7.992936.0468014.29370.25560.35450.29680.4240.24960.13390.21390.54690.65150.720.64260.57930.25560.6515
7.992937.0481014.60870.27370.38090.30780.42790.2680.16450.24590.55580.65080.71090.62080.6120.27370.6508
7.992938.0494014.70620.2390.34280.26850.40640.21520.08710.1960.52410.63370.71360.61090.5630.2390.6337
7.572239.0507014.88260.2520.35640.28010.42030.22440.10910.21450.54090.63990.71450.60590.5880.2520.6399
7.572240.0520014.77450.26770.36710.30170.440.26540.15790.22510.55050.65050.72820.62180.58910.26770.6505
7.572241.0533014.94960.24550.34070.27280.41840.21760.10360.20460.53430.6380.72450.59110.58590.24550.638
7.572242.0546015.11040.23750.33460.2590.41240.20090.0970.18350.51910.63760.71820.61580.56520.23750.6376
7.141143.0559015.15140.24550.34690.26430.41230.23040.09740.20260.52570.6380.71180.60790.58260.24550.638
7.141144.0572015.09400.23830.34230.26130.39930.21160.09840.1990.50990.62080.70180.58220.56630.23830.6208
7.141145.0585015.09880.24590.35220.26920.40060.23350.10530.21650.51910.62410.70360.58420.57280.24590.6241

Framework versions

  • —Transformers 4.52.0.dev0
  • —Pytorch 2.6.0+cu124
  • —Datasets 3.5.0
  • —Tokenizers 0.21.1