metadata
base_model: openai/clip-vit-base-patch32
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: document-spoof-clip
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: imagefolder
type: imagefolder
config: default
split: validation
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.9428571428571428
document-spoof-clip
This model is a fine-tuned version of openai/clip-vit-base-patch32 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2587
- Accuracy: 0.9429
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.8421 | 4 | 0.2112 | 0.9571 |
| No log | 1.8947 | 9 | 0.1227 | 0.9857 |
| 0.295 | 2.9474 | 14 | 0.1203 | 0.9571 |
| 0.295 | 4.0 | 19 | 0.0635 | 0.9714 |
| 0.0962 | 4.8421 | 23 | 0.2939 | 0.9429 |
| 0.0962 | 5.8947 | 28 | 0.2483 | 0.9286 |
| 0.163 | 6.9474 | 33 | 0.0712 | 0.9857 |
| 0.163 | 8.0 | 38 | 0.0474 | 0.9714 |
| 0.0646 | 8.8421 | 42 | 0.2012 | 0.9429 |
| 0.0646 | 9.8947 | 47 | 0.3587 | 0.9 |
| 0.1048 | 10.9474 | 52 | 0.0427 | 0.9857 |
| 0.1048 | 12.0 | 57 | 0.0149 | 0.9857 |
| 0.0519 | 12.8421 | 61 | 0.1616 | 0.9571 |
| 0.0519 | 13.8947 | 66 | 0.2286 | 0.9571 |
| 0.0151 | 14.9474 | 71 | 0.1369 | 0.9571 |
| 0.0151 | 16.0 | 76 | 0.2154 | 0.9571 |
| 0.0455 | 16.8421 | 80 | 0.2587 | 0.9429 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1