Edit model card

cvt-13-finetuned-eurosat

This model is a fine-tuned version of microsoft/cvt-13 on the imagefolder dataset. It achieves the following results on the evaluation set:

  • Loss: 1.0659
  • Accuracy: 0.6947

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: 0.0002
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 6

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.0879 0.9362 11 1.8334 0.4316
1.7897 1.9574 23 1.4727 0.5789
1.5798 2.9787 35 1.2478 0.5895
1.4111 4.0 47 1.1628 0.6211
1.3642 4.9362 58 1.0785 0.6842
1.2403 5.6170 66 1.0659 0.6947

Framework versions

  • Transformers 4.40.0
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1
Downloads last month
9
Safetensors
Model size
19.6M 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 pk3388/cvt-13-finetuned-eurosat

Base model

microsoft/cvt-13
Finetuned
(5)
this model

Evaluation results