If you want more details about the project (like the source code and other details), the developer, and other information (like social profiles and contact info), click the 'About' button on the navigation bar above. Or click the button below:
If you would like to rate this web application, click the 'Rate' button on the navigation bar above, or use the button below:
The profile dropdown contains 3 options: Settings, API documents and Logout. Click the 'Profile' button on the navigation bar above.
The Settings section contains 3 windows: Edit name and password, Edit email and Delete account.
Here the user can change either the name, the password or both. To make these changes click the Edit name and password button of the Settings dropdown at the Profile dropdown, or click here:.
Here the user can change the email, in which the user will receive a verification link to confirm the change at the new email. To make this change click the Edit email button of the Settings dropdown at the Profile dropdown, or click here:
Here the user can delete perminately the account. To make this change click the Delete account button of the Settings dropdown at the Profile dropdown, or click here:.
Here you can find information about the API files used in this web application, a Github link of this project and your token. Go to the Profile dropdown to access the APIs, or click the button below:.
Here the user can Logout of the web application. To logout, click the logout button of the Profile dropdown, or click here:.
The services section contains 2 windows: Create new model and Pretrained models.
In this section, users can select an existing dataset or upload their own. After uploading, they can choose parameters, and execute the k-NN algorithm. Once executed, an evaluation appears, displaying metrics per class, max accuracy and average metrics, and the best parameters. To save the model, enter its name and click 'Save model'.
The picture on top (below this text), shows the half top of the evaluation window, in which includes the 'Metrics per Class' and the 'Accuracy and Average Metrics' tables.
The picture below the top, is the other half of the evaluation window, in which shows the 'Best parameters' table, the text input for the model's name, and the 'Save model' button.
Previously built models appear here. Users can select a pretrained model and apply it to an unclassified dataset to predict class values. If the dataset contains a class column, then this displays the dataset with columns the features the user selected, the class and the predicted class process, and also provides both average and per-class metrics, and the classified dataset can be exported in CSV format. If the dataset does not contain a class column, then it displays only the dataset with columns the features the user selected and the predicted class and the classified dataset can be exported in CSV format.
The picture on top (below this text), shows the dataset table (features + predicted class columns) and the 'Export to .csv' button.
The pictures below the top, is the first half of the classified dataset window, in which includes the dataset table (features + class + predicted class columns).
The second half of the classified dataset window includes the 'Metrics per Class' and the 'Average Metrics' tables (to see the tables, you've to press the respective button) the 'Export to .csv' button.