metadata
license: apache-2.0
base_model: google/vit-hybrid-base-bit-384
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: hybrid-cnn-vit
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: 0.8707767328456983
hybrid-cnn-vit
This model is a fine-tuned version of google/vit-hybrid-base-bit-384 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3384
- Accuracy: 0.8708
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.5277 | 1.0 | 202 | 0.3903 | 0.8210 |
| 0.4623 | 2.0 | 404 | 0.3478 | 0.8415 |
| 0.4497 | 3.0 | 606 | 0.3334 | 0.8520 |
| 0.4074 | 4.0 | 808 | 0.3397 | 0.8460 |
| 0.3552 | 5.0 | 1010 | 0.3227 | 0.8624 |
| 0.3637 | 6.0 | 1212 | 0.3230 | 0.8617 |
| 0.3316 | 7.0 | 1414 | 0.3189 | 0.8673 |
| 0.31 | 8.0 | 1616 | 0.3804 | 0.8492 |
| 0.2324 | 9.0 | 1818 | 0.3382 | 0.8662 |
| 0.234 | 10.0 | 2020 | 0.3384 | 0.8708 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2