<!-- 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. -->
plant-seedlings-model
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2095
- Accuracy: 0.9701
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: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.5298 | 0.8 | 100 | 0.3284 | 0.8816 |
| 0.227 | 1.6 | 200 | 0.2696 | 0.9164 |
| 0.0994 | 2.4 | 300 | 0.4517 | 0.8886 |
| 0.1671 | 3.2 | 400 | 0.3317 | 0.9164 |
| 0.1438 | 4.0 | 500 | 0.2582 | 0.9363 |
| 0.0072 | 4.8 | 600 | 0.2833 | 0.9413 |
| 0.0873 | 5.6 | 700 | 0.2988 | 0.9154 |
| 0.0193 | 6.4 | 800 | 0.2307 | 0.9483 |
| 0.0141 | 7.2 | 900 | 0.3538 | 0.9333 |
| 0.0057 | 8.0 | 1000 | 0.2162 | 0.9552 |
| 0.0046 | 8.8 | 1100 | 0.2587 | 0.9502 |
| 0.0235 | 9.6 | 1200 | 0.3755 | 0.9443 |
| 0.0399 | 10.4 | 1300 | 0.2639 | 0.9493 |
| 0.0031 | 11.2 | 1400 | 0.2164 | 0.9632 |
| 0.0001 | 12.0 | 1500 | 0.2075 | 0.9701 |
| 0.0001 | 12.8 | 1600 | 0.2074 | 0.9701 |
| 0.0001 | 13.6 | 1700 | 0.2078 | 0.9701 |
| 0.0001 | 14.4 | 1800 | 0.2081 | 0.9701 |
| 0.0001 | 15.2 | 1900 | 0.2084 | 0.9701 |
| 0.0001 | 16.0 | 2000 | 0.2088 | 0.9701 |
| 0.0001 | 16.8 | 2100 | 0.2089 | 0.9701 |
| 0.0001 | 17.6 | 2200 | 0.2092 | 0.9701 |
| 0.0001 | 18.4 | 2300 | 0.2095 | 0.9701 |
| 0.0001 | 19.2 | 2400 | 0.2095 | 0.9701 |
| 0.0001 | 20.0 | 2500 | 0.2095 | 0.9701 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
