Disaster Tweets Classification

To explain all concepts that we’ll learn I decided to tackle the following problem:

This competition on Kaggle contains a dataset with information about various tweets. The objective of the competition is to predict which tweets are about real disasters and which are not.

Setting up your Kaggle credentials to work locally

Before downloading the dataset from any competition on Kaggle, you need to set up your Kaggle credentials.

Create API token

First, sign in to your Kaggle account, and then create an API token. To do this, click on your profile picture in the top right corner of the Kaggle website and select “Settings” from the dropdown menu.

On the “Settings” page, you can find the “API” section:

Click on “Create New Token” button:

This action will download a file named kaggle.json to your default download directory. This file contains your Kaggle credentials. If you open the .json file, you will see a code like the following:

{
	"username":"XXXXXX",
	"key":"XXXXXXXXXXXXXXXXXXXXX"
	}

Where:

• username is your Kaggle username.
• key is your API key.

<aside> ⚠️

Don’t share your credentials with anyone. With these credentials, you can gain free access to your account on Kaggle.

</aside>