Hugging Face's logo

gaborcselle
/
font-identifier

Image Classification
Transformers
TensorBoard
Safetensors
English
resnet
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
gaborcselle commited on
Commit
3b3f7c0
•
1 Parent(s): 8f8ac1b

font-identifier

Browse files
Files changed (1) hide show
  1. README.md +71 -47
README.md CHANGED
@@ -1,56 +1,80 @@
1
  ---
2
- language:
3
- - en
4
  tags:
5
- - font-identification
6
- license: mit
7
- widget:
8
- - text: "What font is this?"
9
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/ArchitectsDaughter-Regular_1.png
10
- example_title: Architects Daughter
11
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Arial_Bold_39.png
12
- example_title: Arial Bold
13
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Courier_28.png
14
- example_title: Courier
15
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Helvetica_3.png
16
- example_title: Helvetica
17
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/IBMPlexSans-Regular_25.png
18
- example_title: IBM Plex Sans
19
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Inter-Regular_43.png
20
- example_title: Inter
21
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Lobster-Regular_25.png
22
- example_title: Lobster
23
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Merriweather-Regular_1.png
24
- example_title: Merriweather
25
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Poppins-Regular_22.png
26
- example_title: Poppins
27
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/RobotoMono-Regular_38.png
28
- example_title: Roboto Mono
29
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Times_New_Roman_Bold Italic_26.png
30
- example_title: Times New Roman Bold Italic
31
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Times_New_Roman_Italic_16.png
32
- example_title: Times New Roman Italic
33
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/TitilliumWeb-Regular_5.png
34
- example_title: Titillium Web
35
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Trebuchet_MS_Italic_47.png
36
- example_title: Trebuchet MS Italic
37
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Trebuchet_MS_11.png
38
- example_title: Trebuchet MS
39
- - src: https://huggingface.co/gaborcselle/font-identifier/hf_sample/Verdana_Bold_43.png
40
- example_title: Verdana Bold
41
  ---
42
 
43
- # Font Identifier Project
 
44
 
45
- Tinker project, Nov 8, 2023.
46
 
47
- Follow along:
48
- - [On Pebble.social](https://pebble.social/@gabor/111376050835874755)
49
- - [On Threads.net](https://www.threads.net/@gaborcselle/post/CzZJpJCpxTz)
50
- - [On Twitter](https://twitter.com/gabor/status/1722300841691103467)
51
 
52
- Generate sample images (note this will work only on Mac): [gen_sample_data.py](gen_sample_data.py)
53
 
54
- Arrange test images into test and train: [arrange_train_test_images.py](arrange_train_test_images.py)
55
 
56
- Train a ResNet18 on the data: [train_font_identifier.py](train_font_identifier.py)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license: apache-2.0
3
+ base_model: microsoft/resnet-18
4
  tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - imagefolder
8
+ metrics:
9
+ - accuracy
10
+ model-index:
11
+ - name: font-identifier
12
+ results:
13
+ - task:
14
+ name: Image Classification
15
+ type: image-classification
16
+ dataset:
17
+ name: imagefolder
18
+ type: imagefolder
19
+ config: default
20
+ split: test
21
+ args: default
22
+ metrics:
23
+ - name: Accuracy
24
+ type: accuracy
25
+ value: 0.38979591836734695
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
26
  ---
27
 
28
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
+ should probably proofread and complete it, then remove this comment. -->
30
 
31
+ # font-identifier
32
 
33
+ This model is a fine-tuned version of [microsoft/resnet-18](https://huggingface.co/microsoft/resnet-18) on the imagefolder dataset.
34
+ It achieves the following results on the evaluation set:
35
+ - Loss: 2.5735
36
+ - Accuracy: 0.3898
37
 
38
+ ## Model description
39
 
40
+ More information needed
41
 
42
+ ## Intended uses & limitations
43
+
44
+ More information needed
45
+
46
+ ## Training and evaluation data
47
+
48
+ More information needed
49
+
50
+ ## Training procedure
51
+
52
+ ### Training hyperparameters
53
+
54
+ The following hyperparameters were used during training:
55
+ - learning_rate: 5e-05
56
+ - train_batch_size: 16
57
+ - eval_batch_size: 16
58
+ - seed: 42
59
+ - gradient_accumulation_steps: 4
60
+ - total_train_batch_size: 64
61
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
+ - lr_scheduler_type: linear
63
+ - lr_scheduler_warmup_ratio: 0.1
64
+ - num_epochs: 3
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
+ | 3.5314 | 0.98 | 30 | 3.2829 | 0.2082 |
71
+ | 2.9107 | 1.98 | 61 | 2.6947 | 0.3633 |
72
+ | 2.6604 | 2.93 | 90 | 2.5735 | 0.3898 |
73
+
74
+
75
+ ### Framework versions
76
+
77
+ - Transformers 4.36.0.dev0
78
+ - Pytorch 2.0.0
79
+ - Datasets 2.12.0
80
+ - Tokenizers 0.14.1