File size: 1,301 Bytes
3adb143
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
---
license: apache-2.0
pipeline_tag: image-classification
library_name: transformers
tags:
- dinov2
- image-classification
- fonts
datasets:
- dchen0/font_crops_v3
language:
- en
base_model:
- facebook/dinov2-base
---

# dchen0/font-classifier
Merged DINOv2‑base checkpoint with LoRA weights for font classification.

<!-- 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. -->

This model is a fine-tuned version of [facebook/dinov2-base-imagenet1k-1-layer](https://huggingface.co/facebook/dinov2-base-imagenet1k-1-layer) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 1.3446824550628662
- Accuracy: 0.5903


## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
-learning_rate 1e-4 
-lora_rank 8 
-lora_alpha 16 
-lora_dropout 0.1 
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear

### Framework versions

- PEFT 0.15.2
- Transformers 4.52.4
- Pytorch 2.7.1
- Datasets 3.6.0
- Tokenizers 0.21.1