|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- f1 |
|
|
base_model: facebook/convnext-base-224-22k |
|
|
model-index: |
|
|
- name: flyswot_iiif |
|
|
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. --> |
|
|
|
|
|
# flyswot_iiif |
|
|
|
|
|
This model is a fine-tuned version of [facebook/convnext-base-224-22k](https://huggingface.co/facebook/convnext-base-224-22k) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 6.1280 |
|
|
- F1: 0.0034 |
|
|
|
|
|
## 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: 2e-05 |
|
|
- train_batch_size: 64 |
|
|
- eval_batch_size: 64 |
|
|
- seed: 666 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 8 |
|
|
- mixed_precision_training: Native AMP |
|
|
- label_smoothing_factor: 0.1 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | |
|
|
|:-------------:|:-----:|:-----:|:---------------:|:------:| |
|
|
| 8.5184 | 0.26 | 500 | 7.9280 | 0.0005 | |
|
|
| 7.7409 | 0.52 | 1000 | 7.5824 | 0.0007 | |
|
|
| 7.4649 | 0.78 | 1500 | 7.3841 | 0.0010 | |
|
|
| 7.3285 | 1.04 | 2000 | 7.2652 | 0.0012 | |
|
|
| 7.1404 | 1.3 | 2500 | 7.1559 | 0.0014 | |
|
|
| 7.0322 | 1.56 | 3000 | 7.0551 | 0.0016 | |
|
|
| 6.9197 | 1.82 | 3500 | 6.9449 | 0.0019 | |
|
|
| 6.7822 | 2.09 | 4000 | 6.8773 | 0.0018 | |
|
|
| 6.6506 | 2.35 | 4500 | 6.7980 | 0.0020 | |
|
|
| 6.5811 | 2.61 | 5000 | 6.7382 | 0.0022 | |
|
|
| 6.538 | 2.87 | 5500 | 6.6582 | 0.0022 | |
|
|
| 6.4136 | 3.13 | 6000 | 6.6013 | 0.0024 | |
|
|
| 6.3325 | 3.39 | 6500 | 6.5369 | 0.0024 | |
|
|
| 6.2566 | 3.65 | 7000 | 6.4875 | 0.0025 | |
|
|
| 6.2285 | 3.91 | 7500 | 6.4342 | 0.0027 | |
|
|
| 6.1281 | 4.17 | 8000 | 6.4066 | 0.0027 | |
|
|
| 6.0762 | 4.43 | 8500 | 6.3674 | 0.0027 | |
|
|
| 6.0309 | 4.69 | 9000 | 6.3336 | 0.0027 | |
|
|
| 6.0123 | 4.95 | 9500 | 6.2932 | 0.0030 | |
|
|
| 5.9089 | 5.21 | 10000 | 6.2835 | 0.0029 | |
|
|
| 5.8901 | 5.47 | 10500 | 6.2481 | 0.0030 | |
|
|
| 5.86 | 5.74 | 11000 | 6.2295 | 0.0030 | |
|
|
| 5.8586 | 6.0 | 11500 | 6.2068 | 0.0033 | |
|
|
| 5.7768 | 6.26 | 12000 | 6.1937 | 0.0031 | |
|
|
| 5.7591 | 6.52 | 12500 | 6.1916 | 0.0032 | |
|
|
| 5.7443 | 6.78 | 13000 | 6.1579 | 0.0033 | |
|
|
| 5.7125 | 7.04 | 13500 | 6.1478 | 0.0033 | |
|
|
| 5.6751 | 7.3 | 14000 | 6.1379 | 0.0035 | |
|
|
| 5.6648 | 7.56 | 14500 | 6.1304 | 0.0035 | |
|
|
| 5.6644 | 7.82 | 15000 | 6.1280 | 0.0034 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.17.0.dev0 |
|
|
- Pytorch 1.10.0+cu111 |
|
|
- Datasets 1.18.3 |
|
|
- Tokenizers 0.11.6 |
|
|
|
