|
--- |
|
license: apache-2.0 |
|
base_model: google/vit-large-patch32-224-in21k |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- imagefolder |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: ViTL-32-224-1e4-batch_16_epoch_4_classes_24 |
|
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: 0.9410919540229885 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# ViTL-32-224-1e4-batch_16_epoch_4_classes_24 |
|
|
|
This model is a fine-tuned version of [google/vit-large-patch32-224-in21k](https://huggingface.co/google/vit-large-patch32-224-in21k) on the imagefolder dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3192 |
|
- Accuracy: 0.9411 |
|
|
|
## 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: 0.0002 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 1.3387 | 0.03 | 100 | 1.3149 | 0.7328 | |
|
| 0.7705 | 0.07 | 200 | 0.7867 | 0.8003 | |
|
| 0.5818 | 0.1 | 300 | 0.6799 | 0.8204 | |
|
| 0.537 | 0.14 | 400 | 0.4596 | 0.8836 | |
|
| 0.4053 | 0.17 | 500 | 0.5233 | 0.8592 | |
|
| 0.3401 | 0.21 | 600 | 0.6987 | 0.8032 | |
|
| 0.5161 | 0.24 | 700 | 0.5360 | 0.8405 | |
|
| 0.3592 | 0.28 | 800 | 0.4567 | 0.8664 | |
|
| 0.284 | 0.31 | 900 | 0.3531 | 0.8966 | |
|
| 0.2266 | 0.35 | 1000 | 0.4766 | 0.8678 | |
|
| 0.2876 | 0.38 | 1100 | 0.6849 | 0.8233 | |
|
| 0.3459 | 0.42 | 1200 | 0.4300 | 0.8851 | |
|
| 0.2598 | 0.45 | 1300 | 0.3651 | 0.9052 | |
|
| 0.5085 | 0.49 | 1400 | 0.4353 | 0.8736 | |
|
| 0.4432 | 0.52 | 1500 | 0.4327 | 0.8678 | |
|
| 0.2403 | 0.56 | 1600 | 0.4481 | 0.8736 | |
|
| 0.4616 | 0.59 | 1700 | 0.5625 | 0.8549 | |
|
| 0.244 | 0.63 | 1800 | 0.4537 | 0.8664 | |
|
| 0.4304 | 0.66 | 1900 | 0.4377 | 0.8879 | |
|
| 0.1581 | 0.7 | 2000 | 0.4487 | 0.8851 | |
|
| 0.1273 | 0.73 | 2100 | 0.5803 | 0.8649 | |
|
| 0.1073 | 0.77 | 2200 | 0.4146 | 0.8865 | |
|
| 0.2694 | 0.8 | 2300 | 0.3707 | 0.9080 | |
|
| 0.1699 | 0.84 | 2400 | 0.3477 | 0.9152 | |
|
| 0.2632 | 0.87 | 2500 | 0.4382 | 0.8951 | |
|
| 0.1191 | 0.91 | 2600 | 0.3614 | 0.9095 | |
|
| 0.1634 | 0.94 | 2700 | 0.3786 | 0.9167 | |
|
| 0.1704 | 0.98 | 2800 | 0.4049 | 0.8865 | |
|
| 0.0117 | 1.01 | 2900 | 0.3248 | 0.9080 | |
|
| 0.0522 | 1.04 | 3000 | 0.3518 | 0.9066 | |
|
| 0.179 | 1.08 | 3100 | 0.4117 | 0.9080 | |
|
| 0.0079 | 1.11 | 3200 | 0.4204 | 0.9023 | |
|
| 0.1191 | 1.15 | 3300 | 0.4253 | 0.9066 | |
|
| 0.0444 | 1.18 | 3400 | 0.4485 | 0.9080 | |
|
| 0.2814 | 1.22 | 3500 | 0.4029 | 0.9167 | |
|
| 0.1599 | 1.25 | 3600 | 0.4882 | 0.8937 | |
|
| 0.0156 | 1.29 | 3700 | 0.4070 | 0.9152 | |
|
| 0.2496 | 1.32 | 3800 | 0.3230 | 0.9282 | |
|
| 0.0407 | 1.36 | 3900 | 0.3894 | 0.9167 | |
|
| 0.1122 | 1.39 | 4000 | 0.4924 | 0.8980 | |
|
| 0.0803 | 1.43 | 4100 | 0.4620 | 0.8937 | |
|
| 0.1398 | 1.46 | 4200 | 0.3461 | 0.9109 | |
|
| 0.1072 | 1.5 | 4300 | 0.4346 | 0.9080 | |
|
| 0.0855 | 1.53 | 4400 | 0.3444 | 0.9267 | |
|
| 0.0065 | 1.57 | 4500 | 0.4178 | 0.9023 | |
|
| 0.0143 | 1.6 | 4600 | 0.3257 | 0.9224 | |
|
| 0.041 | 1.64 | 4700 | 0.3396 | 0.9195 | |
|
| 0.0042 | 1.67 | 4800 | 0.3481 | 0.9253 | |
|
| 0.0117 | 1.71 | 4900 | 0.4299 | 0.9037 | |
|
| 0.132 | 1.74 | 5000 | 0.3819 | 0.9195 | |
|
| 0.0223 | 1.78 | 5100 | 0.4280 | 0.9152 | |
|
| 0.0009 | 1.81 | 5200 | 0.4115 | 0.9239 | |
|
| 0.0578 | 1.85 | 5300 | 0.3844 | 0.9267 | |
|
| 0.0014 | 1.88 | 5400 | 0.4024 | 0.9296 | |
|
| 0.002 | 1.92 | 5500 | 0.4511 | 0.9095 | |
|
| 0.0186 | 1.95 | 5600 | 0.3562 | 0.9353 | |
|
| 0.1249 | 1.99 | 5700 | 0.3672 | 0.9253 | |
|
| 0.0615 | 2.02 | 5800 | 0.3567 | 0.9310 | |
|
| 0.0031 | 2.06 | 5900 | 0.3148 | 0.9325 | |
|
| 0.0212 | 2.09 | 6000 | 0.3752 | 0.9267 | |
|
| 0.0008 | 2.12 | 6100 | 0.3394 | 0.9339 | |
|
| 0.0007 | 2.16 | 6200 | 0.3566 | 0.9339 | |
|
| 0.0771 | 2.19 | 6300 | 0.3514 | 0.9310 | |
|
| 0.0007 | 2.23 | 6400 | 0.4172 | 0.9253 | |
|
| 0.0018 | 2.26 | 6500 | 0.4019 | 0.9267 | |
|
| 0.0058 | 2.3 | 6600 | 0.3383 | 0.9368 | |
|
| 0.0032 | 2.33 | 6700 | 0.3362 | 0.9339 | |
|
| 0.0006 | 2.37 | 6800 | 0.3186 | 0.9382 | |
|
| 0.0005 | 2.4 | 6900 | 0.3366 | 0.9382 | |
|
| 0.0006 | 2.44 | 7000 | 0.3802 | 0.9296 | |
|
| 0.0919 | 2.47 | 7100 | 0.4116 | 0.9296 | |
|
| 0.0005 | 2.51 | 7200 | 0.3063 | 0.9425 | |
|
| 0.0004 | 2.54 | 7300 | 0.3466 | 0.9339 | |
|
| 0.0005 | 2.58 | 7400 | 0.3435 | 0.9368 | |
|
| 0.0004 | 2.61 | 7500 | 0.3080 | 0.9411 | |
|
| 0.0016 | 2.65 | 7600 | 0.3310 | 0.9425 | |
|
| 0.0004 | 2.68 | 7700 | 0.3398 | 0.9368 | |
|
| 0.0004 | 2.72 | 7800 | 0.3446 | 0.9353 | |
|
| 0.0004 | 2.75 | 7900 | 0.3294 | 0.9382 | |
|
| 0.1075 | 2.79 | 8000 | 0.3090 | 0.9425 | |
|
| 0.0004 | 2.82 | 8100 | 0.3218 | 0.9382 | |
|
| 0.0004 | 2.86 | 8200 | 0.3160 | 0.9425 | |
|
| 0.0004 | 2.89 | 8300 | 0.3270 | 0.9397 | |
|
| 0.0004 | 2.93 | 8400 | 0.3273 | 0.9397 | |
|
| 0.0003 | 2.96 | 8500 | 0.3184 | 0.9440 | |
|
| 0.0004 | 3.0 | 8600 | 0.3192 | 0.9411 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.3 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
