Downloads · 30 days
6
1% of all-time downloads
vit-base-tour-demo-v5
jungjongho/vit-base-tour-demo-v5
vit-base-tour-demo-v5 is a image classification model from jungjongho. Use it when you need a label for an image. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
607
Public
Repo size
1 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin344 MB · 100%
At a glance
- Task
- Image Classification
- Library
- transformers
- License
- apache-2.0
- Model type
- vit
- Access
- Public
- Created
- Oct 3, 2022
- Updated
- Oct 4, 2022
- SHA
- 914c63b9
Model card
From the Hugging Face model README
<!-- 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. -->
vit-base-tour-demo-v5
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.1467
- Accuracy: 0.4880
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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 3.4573 | 0.13 | 100 | 3.2038 | 0.3334 |
| 2.9547 | 0.27 | 200 | 2.8725 | 0.3672 |
| 2.6093 | 0.4 | 300 | 2.7619 | 0.3954 |
| 2.6212 | 0.54 | 400 | 2.6269 | 0.3942 |
| 2.5063 | 0.67 | 500 | 2.5060 | 0.4211 |
| 2.3113 | 0.81 | 600 | 2.5348 | 0.4201 |
| 2.5702 | 0.94 | 700 | 2.3345 | 0.4502 |
| 2.0479 | 1.08 | 800 | 2.3183 | 0.4484 |
| 1.754 | 1.21 | 900 | 2.2546 | 0.4661 |
| 1.7772 | 1.34 | 1000 | 2.1994 | 0.4794 |
| 1.9276 | 1.48 | 1100 | 2.1672 | 0.4731 |
| 1.6621 | 1.61 | 1200 | 2.1676 | 0.4845 |
| 1.7063 | 1.75 | 1300 | 2.1446 | 0.4806 |
| 1.8655 | 1.88 | 1400 | 2.1121 | 0.4933 |
| 1.4577 | 2.02 | 1500 | 2.0934 | 0.4955 |
| 1.1857 | 2.15 | 1600 | 2.1128 | 0.4906 |
| 1.1684 | 2.28 | 1700 | 2.1218 | 0.4941 |
| 1.3873 | 2.42 | 1800 | 2.1108 | 0.4957 |
| 1.3545 | 2.55 | 1900 | 2.0985 | 0.4992 |
| 0.9789 | 2.69 | 2000 | 2.0997 | 0.4961 |
| 1.1772 | 2.82 | 2100 | 2.1141 | 0.4951 |
| 1.0968 | 2.96 | 2200 | 2.1097 | 0.4922 |
| 0.7883 | 3.09 | 2300 | 2.1170 | 0.5067 |
| 0.7593 | 3.23 | 2400 | 2.1516 | 0.4847 |
| 0.5671 | 3.36 | 2500 | 2.1414 | 0.4925 |
| 0.6442 | 3.49 | 2600 | 2.1498 | 0.4880 |
| 0.516 | 3.63 | 2700 | 2.1442 | 0.4878 |
| 0.6283 | 3.76 | 2800 | 2.1518 | 0.4882 |
| 0.5629 | 3.9 | 2900 | 2.1467 | 0.4880 |
Framework versions
- Transformers 4.22.2
- Pytorch 1.12.1+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1
- Task
- Image Classification
- Library
- transformers
- Type
- vit
- License
- apache-2.0
- Created
- Oct 3, 2022
- Updated
- Oct 4, 2022