| ---
|
| license: apache-2.0
|
| base_model: microsoft/swin-tiny-patch4-window7-224
|
| tags:
|
| - generated_from_trainer
|
| datasets:
|
| - imagefolder
|
| metrics:
|
| - accuracy
|
| - precision
|
| - recall
|
| - f1
|
| model-index:
|
| - name: batch-size16_DFDC_opencv-1FPS_faces-expand20-aligned_unaugmentation
|
| results:
|
| - task:
|
| name: Image Classification
|
| type: image-classification
|
| dataset:
|
| name: imagefolder
|
| type: imagefolder
|
| config: default
|
| split: test
|
| args: default
|
| metrics:
|
| - name: Accuracy
|
| type: accuracy
|
| value: 0.9794553960440362
|
| - name: Precision
|
| type: precision
|
| value: 0.9849190255339879
|
| - name: Recall
|
| type: recall
|
| value: 0.9907155000951158
|
| - name: F1
|
| type: f1
|
| value: 0.9878087594408504
|
| ---
|
|
|
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| should probably proofread and complete it, then remove this comment. -->
|
|
|
| # batch-size16_DFDC_opencv-1FPS_faces-expand20-aligned_unaugmentation
|
|
|
| This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| It achieves the following results on the evaluation set:
|
| - Loss: 0.0532
|
| - Accuracy: 0.9795
|
| - Precision: 0.9849
|
| - Recall: 0.9907
|
| - F1: 0.9878
|
| - Roc Auc: 0.9964
|
|
|
| ## 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: 16
|
| - eval_batch_size: 16
|
| - seed: 42
|
| - gradient_accumulation_steps: 4
|
| - total_train_batch_size: 64
|
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| - lr_scheduler_type: linear
|
| - lr_scheduler_warmup_ratio: 0.1
|
| - num_epochs: 1
|
|
|
| ### Training results
|
|
|
| | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Roc Auc |
|
| |:-------------:|:------:|:-----:|:---------------:|:--------:|:---------:|:------:|:------:|:-------:|
|
| | 0.0543 | 1.0000 | 18673 | 0.0532 | 0.9795 | 0.9849 | 0.9907 | 0.9878 | 0.9964 |
|
|
|
|
|
| ### Framework versions
|
|
|
| - Transformers 4.41.2
|
| - Pytorch 2.3.1
|
| - Datasets 2.20.0
|
| - Tokenizers 0.19.1
|
|
|
