greathero's picture
Training in progress epoch 18
9bda1d2
|
raw
history blame
9.36 kB
metadata
license: other
base_model: nvidia/mit-b0
tags:
  - generated_from_keras_callback
model-index:
  - name: greathero/mit-b0-finetuned-contrails-morethanx35newercontrailsdataset
    results: []

greathero/mit-b0-finetuned-contrails-morethanx35newercontrailsdataset

This model is a fine-tuned version of nvidia/mit-b0 on an unknown dataset. It achieves the following results on the evaluation set:

  • Train Loss: 0.0048
  • Validation Loss: 0.0047
  • Validation Mean Iou: 0.6598
  • Validation Mean Accuracy: 0.6878
  • Validation Overall Accuracy: 0.9984
  • Validation Accuracy Unlabeled: 0.8252
  • Validation Accuracy Notlabeled: 0.9998
  • Validation Accuracy Otherclass: 0.7809
  • Validation Accuracy Contrail: 0.1453
  • Validation Iou Unlabeled: 0.8013
  • Validation Iou Notlabeled: 0.9984
  • Validation Iou Otherclass: 0.7095
  • Validation Iou Contrail: 0.1298
  • Epoch: 18

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:

  • optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': 6e-05, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
  • training_precision: float32

Training results

Train Loss Validation Loss Validation Mean Iou Validation Mean Accuracy Validation Overall Accuracy Validation Accuracy Unlabeled Validation Accuracy Notlabeled Validation Accuracy Otherclass Validation Accuracy Contrail Validation Iou Unlabeled Validation Iou Notlabeled Validation Iou Otherclass Validation Iou Contrail Epoch
0.2395 0.0367 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 0
0.0286 0.0204 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 1
0.0194 0.0156 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 2
0.0157 0.0144 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 3
0.0141 0.0137 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 4
0.0133 0.0132 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 5
0.0126 0.0125 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 6
0.0118 0.0117 0.2496 0.25 0.9983 0.0 1.0 0.0 0.0 0.0 0.9983 0.0 0.0 7
0.0112 0.0110 0.2551 0.2555 0.9983 0.0 1.0000 0.0222 0.0 0.0 0.9983 0.0220 0.0 8
0.0106 0.0103 0.2744 0.2756 0.9983 0.0180 1.0000 0.0832 0.0013 0.0179 0.9983 0.0801 0.0013 9
0.0100 0.0094 0.3172 0.3225 0.9983 0.0527 1.0000 0.2344 0.0030 0.0521 0.9983 0.2154 0.0030 10
0.0091 0.0087 0.3117 0.3138 0.9983 0.1026 1.0000 0.1470 0.0055 0.1014 0.9983 0.1415 0.0055 11
0.0083 0.0076 0.3844 0.3941 0.9983 0.1540 1.0000 0.4189 0.0037 0.1516 0.9983 0.3840 0.0037 12
0.0073 0.0070 0.4402 0.4781 0.9983 0.1997 0.9999 0.6616 0.0511 0.1978 0.9983 0.5167 0.0479 13
0.0067 0.0062 0.5049 0.5454 0.9983 0.4868 1.0000 0.6893 0.0057 0.4702 0.9983 0.5456 0.0056 14
0.0060 0.0056 0.5133 0.5326 0.9983 0.4799 1.0000 0.6283 0.0221 0.4722 0.9983 0.5608 0.0218 15
0.0055 0.0056 0.5975 0.6504 0.9984 0.7018 0.9999 0.8058 0.0942 0.6731 0.9984 0.6317 0.0868 16
0.0051 0.0050 0.5800 0.6003 0.9984 0.6893 0.9999 0.6644 0.0476 0.6707 0.9984 0.6048 0.0462 17
0.0048 0.0047 0.6598 0.6878 0.9984 0.8252 0.9998 0.7809 0.1453 0.8013 0.9984 0.7095 0.1298 18

Framework versions

  • Transformers 4.35.2
  • TensorFlow 2.15.0
  • Datasets 2.15.0
  • Tokenizers 0.15.0