Exercises

Set 1

1. Go to the Kaggle API documentation and make a Kaggle account. On your profile page you will be able to make an api key, do so and place it in you .env folder. Your goal is to download this file. Note the following:

   • This api uses basic auth and your api token will just have a username and password in it. The syntax for using this with the requests module is requests.get(url, auth=(username, password)) where your key is the password.
   • The "user slug" and "dataset slug" can both be obtained from the page for the file, it is difficult to find from the API
   • The result of this api call will be the raw file data, not a json object. The "content" attribute of the Response object you get from your request can be used to access this data, and you can write it to disk using a similar construct to that used to write the google token to file.
   • Python's default text encoding is "whatever the current user's OS defaults are", which means that using the default encoding reduces compatibility and causes errors. When opening a file, it's important to set the "encoding" parameter, which is almost always "utf-8". Additionally, data sent over a network is often encoded as a stream of bytes, such as the data received for the csv files output by the Kaggle API. bytes objects like these have a .decode() method which can be passed an encoding (such as "utf-8") to decode it into a python string.

2. Once you have the file downloaded, read this short pandas tutorial and read the csv into a dataframe. Eliminate all rows from the dataframe except those for movies released before 2016, and then print their titles.

3. Use the framework we developed to read this spreadsheet into memory and then make it into a dataframe. Confirm that they contain the same data.