metadata
base_model: microsoft/swin-tiny-patch4-window7-224
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
- recall
- precision
- f1
model-index:
- name: FFPP-Raw_1FPS_faces-expand-40-aligned_0Real-1Fake
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: 0.9807551850864278
- name: Recall
type: recall
value: 0.992948461549857
- name: Precision
type: precision
value: 0.982642095849643
- name: F1
type: f1
value: 0.9877683953018849
FFPP-Raw_1FPS_faces-expand-40-aligned_0Real-1Fake
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0514
- Accuracy: 0.9808
- Recall: 0.9929
- Precision: 0.9826
- F1: 0.9878
- Roc Auc: 0.9980
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: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | Precision | F1 | Roc Auc |
---|---|---|---|---|---|---|---|---|
0.1725 | 1.0 | 1348 | 0.1299 | 0.9464 | 0.9628 | 0.9685 | 0.9656 | 0.9850 |
0.1387 | 2.0 | 2696 | 0.0704 | 0.9734 | 0.9854 | 0.9807 | 0.9831 | 0.9959 |
0.0796 | 3.0 | 4044 | 0.0514 | 0.9808 | 0.9929 | 0.9826 | 0.9878 | 0.9980 |
Framework versions
- Transformers 4.39.2
- Pytorch 2.2.2
- Datasets 2.18.0
- Tokenizers 0.15.2