metadata
license: mit
base_model: shi-labs/nat-mini-in1k-224
tags:
- generated_from_trainer
datasets:
- image_folder
metrics:
- accuracy
- f1
model-index:
- name: nat-mini-in1k-224-finetuned-breakhis
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: image_folder
type: image_folder
config: default
split: train
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.9669421487603306
- name: F1
type: f1
value: 0.9612429172231991
nat-mini-in1k-224-finetuned-breakhis
This model is a fine-tuned version of shi-labs/nat-mini-in1k-224 on the image_folder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0983
- Accuracy: 0.9669
- F1: 0.9612
- Roc Auc: 0.9648
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: 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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Roc Auc |
|---|---|---|---|---|---|---|
| 0.3247 | 0.99 | 59 | 0.2084 | 0.9157 | 0.8968 | 0.8836 |
| 0.1338 | 2.0 | 119 | 0.1686 | 0.9355 | 0.9266 | 0.9437 |
| 0.1078 | 2.99 | 178 | 0.0986 | 0.9694 | 0.9636 | 0.9597 |
| 0.0795 | 4.0 | 238 | 0.0957 | 0.9719 | 0.9668 | 0.9660 |
| 0.0522 | 4.96 | 295 | 0.0983 | 0.9669 | 0.9612 | 0.9648 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.1.2
- Datasets 2.1.0
- Tokenizers 0.15.2