Provided several datasets about Pokemon species data about the history of matches between Pokemon. File pokemon.csv contains complete data of 800 Pokemon species, while file combats.csv contains historical data of the Pokemon duel with the winner. Source: click here.
Dengan dataset tersebut, buatlah sebuah aplikasi Flask berisi model machine learning (model bebas) yang dapat memprediksi pemenang dari duel Pokemon. Aplikasi yang dibuat harus memenuhi syarat minimal berikut:
With this dataset, I created a Flask application containing a Gradient Boost Classifier Machine Learning model that can predict the winner of a Pokemon duel. Following below are the description:
-
The application server will run on localhost: 2070 and when the user makes a GET request via the browser a simple HTML page will appear containing 2 inputs text and a button. Users can enter the name of the Pokemon that will be contested in the available text input. When the user presses the 'Battle!' button, the application will process the data entered by the user.
-
If the data is successfully processed, then the user will be redirect to localhost: 2070 / result containing a HTML page that displays:
-
Pokemon Picture
- Use Poke API (click here)
- GET to (click here)
-
Pokemon skill comparison chart (from the dataset):
- HP
- Attack
- Defense
- Special Attack
- Special Defense
- Speed
-
Possible Winners along with percentage (%) probability
This page also includes a button to return to the start page. The design of HTML display doesn't have to be the same as the example problem, prioritize features! Example:
-
-
But if the data is not found, there is not in the dataset. the user enters a url that is not available, then the user will be redirected to the HTML page which provides information that the data is not found. This page also includes a button to return to the start page.
-
And if the user enters a url that is not available, then the user will be redirected to the HTML page which provides information ERROR 404. This page also includes a button to return to the start page.
Note: For further discussion if you have questions, criticisms, and suggestions, feel free to contact my email: [email protected].