RohitManglik commited on
Commit
9fa8bf7
·
verified ·
1 Parent(s): af24760

Delete README.txt

Browse files
Files changed (1) hide show
  1. README.txt +0 -219
README.txt DELETED
@@ -1,219 +0,0 @@
1
- ---
2
- language:
3
- - en
4
-
5
- license: cc-by-4.0
6
-
7
- pretty_name: InfoBay.AI SVG Dataset Catalogue
8
-
9
- size_categories:
10
- - 100K<n<1M
11
-
12
- tags:
13
- - svg
14
- - vector
15
- - icons
16
- - ui
17
- - ux
18
- - branding
19
- - logos
20
- - illustrations
21
- - graphic-design
22
- - computer-vision
23
- - multimodal
24
- - generative-ai
25
- - frontend-ai
26
- - enterprise-ai
27
- - design-automation
28
- dataset_info:
29
- features:
30
- - name: image
31
- dtype: image
32
- - name: filename
33
- dtype: string
34
- - name: svg
35
- dtype: string
36
-
37
- configs:
38
- - config_name: default
39
- data_files:
40
- - split: train
41
- path: "train-*.parquet"
42
- ---
43
-
44
- # InfoBay.AI SVG Dataset Catalogue
45
-
46
- ## Overview
47
-
48
- The **InfoBay.AI SVG Dataset Catalogue** is a professionally curated collection of **873,754 scalable vector graphics (SVG)** designed for Artificial Intelligence, Computer Vision, UI/UX Design, Search Systems, Frontend Development, Design Automation, and Multimodal Foundation Models.
49
-
50
- The dataset contains high-quality vector assets spanning user interface components, branding materials, logos, technology icons, healthcare graphics, business illustrations, communication symbols, and media-related visual elements.
51
-
52
- For efficient storage, cloud-native distribution, and scalable AI workflows, the original SVG files have been serialized into **Apache Parquet** while preserving the original filenames, directory structure, file extensions, and binary content. Every original SVG asset can be reconstructed exactly from the provided Parquet files.
53
-
54
- ---
55
-
56
- # Dataset Statistics
57
-
58
- | Property | Value |
59
- |----------|------:|
60
- | Dataset Name | InfoBay.AI SVG Dataset Catalogue |
61
- | Asset Type | SVG (Scalable Vector Graphics) |
62
- | Total Assets | 873,754 |
63
- | Storage Format | Apache Parquet |
64
- | License | CC BY 4.0 |
65
-
66
- ---
67
-
68
- # Dataset Categories
69
-
70
- The collection includes vector graphics across multiple industries and applications.
71
-
72
- ### Key Categories
73
-
74
- - UI Controls
75
- - Logos & Branding
76
- - Technology & Devices
77
- - Communication
78
- - Business & Finance
79
- - Health & Medical
80
- - Media & Entertainment
81
- - Icons
82
- - Web Components
83
- - Interface Elements
84
- - Navigation Assets
85
- - Product Illustrations
86
- - Educational Graphics
87
- - Marketing Assets
88
-
89
- ---
90
-
91
- # Primary AI Applications
92
-
93
- This dataset is suitable for:
94
-
95
- - Generative AI
96
- - Computer Vision
97
- - Vision Language Models (VLMs)
98
- - Large Multimodal Models (LMMs)
99
- - UI Component Recognition
100
- - Frontend Code Generation
101
- - Design Retrieval
102
- - Image Similarity Search
103
- - Search Indexing
104
- - Vector Embedding Models
105
- - OCR Research
106
- - Design Recommendation
107
- - Interface Understanding
108
- - Human-Computer Interaction
109
-
110
- ---
111
-
112
- # Enterprise Applications
113
-
114
- - UI / UX Design
115
- - Frontend Development
116
- - Graphic Design
117
- - Brand Asset Management
118
- - Enterprise Search
119
- - Marketing Automation
120
- - Product Design
121
- - Design Systems
122
- - Digital Asset Management
123
- - AI-assisted Creative Platforms
124
-
125
- ---
126
-
127
- # Storage Format
128
-
129
- The dataset is distributed in **Apache Parquet** format.
130
-
131
- Each row represents one original SVG asset.
132
-
133
- Schema:
134
-
135
- | Column | Description |
136
- |---------|-------------|
137
- | relative_path | Original file path |
138
- | filename | Original filename |
139
- | extension | File extension |
140
- | size | Original file size (bytes) |
141
- | content | Binary SVG content |
142
-
143
- The complete directory structure and SVG files can be restored programmatically from the Parquet dataset.
144
-
145
- ---
146
-
147
- # Why Apache Parquet?
148
-
149
- Apache Parquet provides:
150
-
151
- - Efficient compression
152
- - Faster loading
153
- - Reduced storage requirements
154
- - Hugging Face compatibility
155
- - Streaming support
156
- - Cloud-native architecture
157
- - Distributed processing
158
- - Scalable AI pipelines
159
-
160
- ---
161
-
162
- # Loading the Dataset
163
-
164
- ```python
165
- from datasets import load_dataset
166
-
167
- dataset = load_dataset("YOUR_USERNAME/YOUR_DATASET_NAME")
168
-
169
- print(dataset)
170
- ```
171
-
172
- ---
173
-
174
- # Example
175
-
176
- ```python
177
- sample = dataset["train"][0]
178
-
179
- print(sample["filename"])
180
- print(sample["extension"])
181
- print(sample["size"])
182
- print(sample["content"])
183
- ```
184
-
185
- ---
186
-
187
-
188
- # Citation
189
-
190
- ```bibtex
191
- @dataset{infobay_svg_dataset_2026,
192
- title = {InfoBay.AI SVG Dataset Catalogue},
193
- author = {InfoBay.AI},
194
- year = {2026},
195
- publisher = {InfoBay.AI},
196
- license = {CC BY 4.0}
197
- }
198
- ```
199
-
200
- ---
201
-
202
-
203
-
204
- **Data Creation**
205
-
206
- Procured through formal agreements and generated in the ordinary course of business.
207
-
208
- **Considerations**
209
-
210
-
211
- This dataset is provided for research and educational purposes only. It contains only sample data. For access to the full dataset and enterprise licensing options, please visit our website InfoBay.AI or contact us directly.
212
-
213
- -Ph: (+91) 9236398619
214
- -Email: datareq@infobay.ai
215
- ---
216
-
217
- # Acknowledgements
218
-
219
- The **InfoBay.AI SVG Dataset Catalogue** has been curated to accelerate research in Computer Vision, UI/UX Intelligence, Design Automation, Frontend AI, Multimodal Learning, Enterprise Search, and Generative AI. We hope this collection contributes to the development of next-generation intelligent design systems.