|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: convnextv2-base-22k-224-finetuned-critique-100k |
|
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. --> |
|
|
|
# convnextv2-base-22k-224-finetuned-critique-100k |
|
|
|
This model is a fine-tuned version of [facebook/convnextv2-base-22k-224](https://huggingface.co/facebook/convnextv2-base-22k-224) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1320 |
|
- Accuracy: 0.9479 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 128 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 0.6277 | 0.07 | 50 | 0.5987 | 0.6767 | |
|
| 0.5459 | 0.14 | 100 | 0.5187 | 0.7401 | |
|
| 0.4397 | 0.21 | 150 | 0.4448 | 0.7768 | |
|
| 0.4197 | 0.28 | 200 | 0.3686 | 0.8401 | |
|
| 0.3397 | 0.36 | 250 | 0.3153 | 0.8664 | |
|
| 0.3345 | 0.43 | 300 | 0.3071 | 0.8701 | |
|
| 0.3177 | 0.5 | 350 | 0.2576 | 0.8938 | |
|
| 0.3182 | 0.57 | 400 | 0.2546 | 0.8926 | |
|
| 0.2596 | 0.64 | 450 | 0.2320 | 0.9004 | |
|
| 0.2563 | 0.71 | 500 | 0.2205 | 0.9082 | |
|
| 0.2543 | 0.78 | 550 | 0.2142 | 0.9147 | |
|
| 0.2768 | 0.85 | 600 | 0.2136 | 0.9132 | |
|
| 0.2486 | 0.92 | 650 | 0.2052 | 0.9175 | |
|
| 0.2504 | 1.0 | 700 | 0.2314 | 0.9058 | |
|
| 0.2437 | 1.07 | 750 | 0.1943 | 0.9235 | |
|
| 0.212 | 1.14 | 800 | 0.2019 | 0.9183 | |
|
| 0.1891 | 1.21 | 850 | 0.1845 | 0.9254 | |
|
| 0.2105 | 1.28 | 900 | 0.1834 | 0.9288 | |
|
| 0.2285 | 1.35 | 950 | 0.1994 | 0.9206 | |
|
| 0.2214 | 1.42 | 1000 | 0.1804 | 0.9251 | |
|
| 0.1848 | 1.49 | 1050 | 0.1975 | 0.9196 | |
|
| 0.191 | 1.56 | 1100 | 0.1795 | 0.9269 | |
|
| 0.1794 | 1.64 | 1150 | 0.1606 | 0.9358 | |
|
| 0.2084 | 1.71 | 1200 | 0.1807 | 0.9293 | |
|
| 0.199 | 1.78 | 1250 | 0.1697 | 0.9307 | |
|
| 0.1874 | 1.85 | 1300 | 0.1650 | 0.9372 | |
|
| 0.1681 | 1.92 | 1350 | 0.1515 | 0.939 | |
|
| 0.1696 | 1.99 | 1400 | 0.1473 | 0.9416 | |
|
| 0.1651 | 2.06 | 1450 | 0.1489 | 0.9428 | |
|
| 0.1627 | 2.13 | 1500 | 0.1529 | 0.9395 | |
|
| 0.1754 | 2.2 | 1550 | 0.1540 | 0.9379 | |
|
| 0.1302 | 2.28 | 1600 | 0.1579 | 0.939 | |
|
| 0.1643 | 2.35 | 1650 | 0.1518 | 0.9401 | |
|
| 0.1938 | 2.42 | 1700 | 0.1479 | 0.941 | |
|
| 0.1441 | 2.49 | 1750 | 0.1451 | 0.9436 | |
|
| 0.1478 | 2.56 | 1800 | 0.1324 | 0.9472 | |
|
| 0.1275 | 2.63 | 1850 | 0.1340 | 0.9466 | |
|
| 0.1582 | 2.7 | 1900 | 0.1501 | 0.9391 | |
|
| 0.1472 | 2.77 | 1950 | 0.1354 | 0.9451 | |
|
| 0.1522 | 2.84 | 2000 | 0.1309 | 0.9479 | |
|
| 0.1593 | 2.92 | 2050 | 0.1433 | 0.9452 | |
|
| 0.1541 | 2.99 | 2100 | 0.1381 | 0.9466 | |
|
| 0.1297 | 3.06 | 2150 | 0.1320 | 0.9479 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.30.2 |
|
- Pytorch 1.13.1 |
|
- Datasets 2.13.1 |
|
- Tokenizers 0.13.3 |
|
|
