|
--- |
|
license: other |
|
base_model: apple/mobilevitv2-1.0-imagenet1k-256 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: quickdraw-MobileVITV2-1.0-Finetune |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# quickdraw-MobileVITV2-1.0-Finetune |
|
|
|
This model is a fine-tuned version of [apple/mobilevitv2-1.0-imagenet1k-256](https://huggingface.co/apple/mobilevitv2-1.0-imagenet1k-256) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0138 |
|
- Accuracy: 0.7524 |
|
|
|
## 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: 0.0008 |
|
- train_batch_size: 512 |
|
- eval_batch_size: 512 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 10000 |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:------:|:-----:|:---------------:|:--------:| |
|
| 1.4934 | 0.5688 | 5000 | 1.4418 | 0.6444 | |
|
| 1.2717 | 1.1377 | 10000 | 1.2881 | 0.6771 | |
|
| 1.1742 | 1.7065 | 15000 | 1.1661 | 0.7052 | |
|
| 1.0846 | 2.2753 | 20000 | 1.1149 | 0.7178 | |
|
| 1.0619 | 2.8441 | 25000 | 1.0778 | 0.7261 | |
|
| 1.0029 | 3.4130 | 30000 | 1.0556 | 0.7322 | |
|
| 0.9936 | 3.9818 | 35000 | 1.0317 | 0.7375 | |
|
| 0.9429 | 4.5506 | 40000 | 1.0150 | 0.7424 | |
|
| 0.8818 | 5.1195 | 45000 | 1.0119 | 0.7451 | |
|
| 0.8868 | 5.6883 | 50000 | 0.9947 | 0.7486 | |
|
| 0.8323 | 6.2571 | 55000 | 1.0007 | 0.7491 | |
|
| 0.838 | 6.8259 | 60000 | 0.9854 | 0.7522 | |
|
| 0.7835 | 7.3948 | 65000 | 0.9989 | 0.7521 | |
|
| 0.7836 | 7.9636 | 70000 | 0.9900 | 0.7535 | |
|
| 0.7451 | 8.5324 | 75000 | 1.0044 | 0.7529 | |
|
| 0.7207 | 9.1013 | 80000 | 1.0054 | 0.7531 | |
|
| 0.721 | 9.6701 | 85000 | 1.0081 | 0.7529 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.2 |
|
- Pytorch 2.2.1 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
