Update README.md
Browse files
README.md
CHANGED
|
@@ -5,5 +5,37 @@ pipeline_tag: image-classification
|
|
| 5 |
license: mit
|
| 6 |
language:
|
| 7 |
- en
|
| 8 |
-
library_name:
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
license: mit
|
| 6 |
language:
|
| 7 |
- en
|
| 8 |
+
library_name: transformers
|
| 9 |
+
tags:
|
| 10 |
+
- dinov2
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# dchen0/font-classifier
|
| 14 |
+
Merged DINOv2‑base checkpoint with LoRA weights for font classification.
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
## Training procedure
|
| 21 |
+
|
| 22 |
+
### Training hyperparameters
|
| 23 |
+
|
| 24 |
+
The following hyperparameters were used during training:
|
| 25 |
+
-learning_rate 1e-4
|
| 26 |
+
-lora_rank 8
|
| 27 |
+
-lora_alpha 16
|
| 28 |
+
-lora_dropout 0.1
|
| 29 |
+
- train_batch_size: 32
|
| 30 |
+
- eval_batch_size: 32
|
| 31 |
+
- seed: 42
|
| 32 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 33 |
+
- lr_scheduler_type: linear
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- PEFT 0.15.2
|
| 38 |
+
- Transformers 4.52.4
|
| 39 |
+
- Pytorch 2.7.1
|
| 40 |
+
- Datasets 3.6.0
|
| 41 |
+
- Tokenizers 0.21.1
|