metadata

library_name: transformers
license: apache-2.0
base_model: microsoft/swin-base-patch4-window7-224
tags:
  - generated_from_trainer
datasets:
  - kedimestan/RetinoblastomaHealthyImageDataset
metrics:
  - accuracy
model-index:
  - name: swin-base-patch4-window7-224
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 1

swin-base-patch4-window7-224

Bu model, microsoft/swin-base-patch4-window7-224 modelin retinoblastoma veri seti için ince ayar yapılmış versiyonudur. Test veri setindeki sonuçları aşağıdaki gibidir:

Loss: 0.3809
Accuracy: 1.0

Model description

Swin tabanlı bir modeldir. 5e-5 öğrenme oranı ile 10 epoch ile eğitim yapıldı. Optimizasyon için Adam kullanıldı. veri seti %80-%20 olacak şekilde sırasyıla eğitim ve test verisetine bölündü.

The following hyperparameters were used during training:

learning_rate: 5e-05
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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.5714	1	0.8119	0.3913
No log	1.7143	3	0.6544	0.5217
No log	2.8571	5	0.3809	1.0
No log	4.0	7	0.2352	1.0
No log	4.5714	8	0.1875	1.0
0.4988	5.7143	10	0.1481	1.0

Framework versions

Transformers 4.44.2
Pytorch 2.4.1+cu121
Datasets 3.2.0
Tokenizers 0.19.1