Birdsclassification

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3057
  • Accuracy: 0.9307

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: 0.0003
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 256
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 16
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
5.42 1.0 262 3.6698 0.7571
1.7968 2.0 525 0.9179 0.8396
0.6598 3.0 787 0.6370 0.8654
0.4867 4.0 1050 0.5493 0.8765
0.4055 5.0 1312 0.5093 0.8833
0.3513 6.0 1575 0.4602 0.8892
0.3053 7.0 1837 0.4350 0.8977
0.2692 8.0 2100 0.4130 0.9021
0.2446 9.0 2362 0.4218 0.9018
0.2267 10.0 2625 0.3667 0.9130
0.2018 11.0 2887 0.3632 0.9154
0.1842 12.0 3150 0.3533 0.9154
0.1636 13.0 3412 0.3396 0.9206
0.1511 14.0 3675 0.3125 0.9266
0.1411 15.0 3937 0.2833 0.9329
0.1259 15.97 4192 0.3057 0.9307

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
19
Safetensors
Model size
86.2M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for Hemg/Birdsclassification

Finetuned
(1976)
this model