Edit model card

resnet-50-finetuned-student_two_classes

This model is a fine-tuned version of microsoft/resnet-50 on the imagefolder dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4531
  • Accuracy: 0.85

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
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.5955 1.0 13 0.4665 0.85
0.5303 2.0 26 0.4790 0.85
0.6127 3.0 39 0.4787 0.85
0.5025 4.0 52 0.4547 0.85
0.471 5.0 65 0.4621 0.85
0.4673 6.0 78 0.4775 0.86
0.4492 7.0 91 0.4648 0.86
0.4144 8.0 104 0.4733 0.85
0.4963 9.0 117 0.4575 0.85
0.4149 10.0 130 0.4691 0.85
0.4588 11.0 143 0.4596 0.84
0.3995 12.0 156 0.4754 0.85
0.359 13.0 169 0.4616 0.85
0.4246 14.0 182 0.4552 0.85
0.4001 15.0 195 0.4839 0.85
0.3919 16.0 208 0.4708 0.85
0.4137 17.0 221 0.4416 0.85
0.3912 18.0 234 0.4507 0.85
0.4322 19.0 247 0.4237 0.85
0.4043 20.0 260 0.4531 0.85

Framework versions

  • Transformers 4.40.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1
Downloads last month
13
Safetensors
Model size
23.6M params
Tensor type
F32
·

Finetuned from

Evaluation results