📒 image-classification-model
This model has undergone training on the "image-classification" dataset, focusing on multi-class classification to categorize specific segments of websites. Each segment corresponds to one of six potential features, encompassing a broad spectrum of web elements, including:
Button: Identifying interactive buttons that users can click or tap on for various website functions.
Textfield: Recognizing text input fields where users can type or enter information.
Checkbox: Detecting checkboxes that users can select or deselect to make choices or indicate preferences.
Radiobutton: Identifying radio buttons that allow users to choose a single option from a list.
Tables: Recognizing tabular data structures that organize information in rows and columns.
AppBar: Detecting app bars or navigation bars typically found at the top of web pages, often containing menus, search bars, or branding elements.
This extensive training equips the model with the ability to accurately classify these web elements.
🧪 Dataset Content
The dataset is structured to facilitate the analysis of website components. It includes various types of objects commonly found on websites, such as buttons, text fields, checkboxes, radio buttons, tables, and app bars. Each object type is organized into its respective category within the dataset, allowing for precise classification.
| Web Element Category | Quantity of images |
|---|---|
| Button | 2934 |
| Textfield | 100 |
| Checkbox | 422 |
| Radiobutton | 466 |
| Tables | 100 |
| AppBar | 100 |
🤗 Model Trained Using AutoTrain
- Problem type: Multi-class Classification
- Model ID: 86974143294
- CO2 Emissions (in grams): 1.6524
📐 Validation Metrics
- Loss: 0.079
- Accuracy: 0.983
- Macro F1: 0.967
- Micro F1: 0.983
- Weighted F1: 0.983
- Macro Precision: 0.971
- Micro Precision: 0.983
- Weighted Precision: 0.983
- Macro Recall: 0.964
- Micro Recall: 0.983
- Weighted Recall: 0.983
- Downloads last month
- 1