detr-r50-finetuned-mist1-gb-8ah-6l

This model is a fine-tuned version of polejowska/detr-r50-cd45rb-8ah-6l on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.9224

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: 1e-05
  • train_batch_size: 4
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 50
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
2.5222 1.0 115 2.2563
2.3827 2.0 230 2.2211
2.3441 3.0 345 2.2602
2.2896 4.0 460 2.2359
2.2828 5.0 575 2.2431
2.2972 6.0 690 2.1629
2.3007 7.0 805 2.1545
2.2951 8.0 920 2.1153
2.2595 9.0 1035 2.1553
2.2327 10.0 1150 2.2060
2.2023 11.0 1265 2.0452
2.2117 12.0 1380 2.0879
2.1805 13.0 1495 2.1812
2.1344 14.0 1610 2.0992
2.1057 15.0 1725 1.9834
2.086 16.0 1840 1.9610
2.0591 17.0 1955 2.1007
2.053 18.0 2070 2.0561
2.0387 19.0 2185 2.0596
2.0161 20.0 2300 1.9885
2.0374 21.0 2415 2.0041
2.0233 22.0 2530 2.0103
2.0363 23.0 2645 2.0541
1.9837 24.0 2760 1.9924
1.9943 25.0 2875 2.0558
1.9846 26.0 2990 1.9874
1.9601 27.0 3105 1.9554
1.9837 28.0 3220 1.9989
1.9664 29.0 3335 1.9876
1.966 30.0 3450 1.9755
1.9226 31.0 3565 1.9357
1.9405 32.0 3680 1.9240
1.9035 33.0 3795 1.9411
1.8924 34.0 3910 1.9291
1.8801 35.0 4025 1.9661
1.8698 36.0 4140 1.9105
1.8572 37.0 4255 1.9448
1.8756 38.0 4370 1.9675
1.8593 39.0 4485 1.9365
1.8713 40.0 4600 1.9383
1.8436 41.0 4715 1.9671
1.83 42.0 4830 1.9527
1.857 43.0 4945 1.9448
1.8318 44.0 5060 1.9366
1.8177 45.0 5175 1.9389
1.8034 46.0 5290 1.9050
1.8226 47.0 5405 1.9226
1.818 48.0 5520 1.9150
1.8148 49.0 5635 1.9169
1.7984 50.0 5750 1.9224

Framework versions

  • Transformers 4.35.0
  • Pytorch 2.0.0
  • Datasets 2.1.0
  • Tokenizers 0.14.1
Downloads last month
16
Safetensors
Model size
41.6M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for polejowska/detr-r50-finetuned-mist1-gb-8ah-6l

Finetuned
(2)
this model