Python in general supports many type of raw files and sources through libraries like `pandas`.

`Pandas` has many helpful `read_filetype()` functions to handle many file types, for example:

`read_csv()`
`read_excel()`
`read_json()`
`read_html()`
`read_sql()`
`read_pickle()`

> Note
>
> See docs for detailed info: https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html

In our example, the training data is in csv format and is stored in `"/kaggle/input/indian-food-101/indian_food.csv"`. We will use `read_csv()` function, it accepts a filepath parameter.

The output is a DataFrame called `IndianFoods`.

DataFrame
- Pandas specific Data structure, to store data in tabular format;
- Looks similar to SQL table;
- Has a lot of associated functions, similar to table-level commands in SQL (SELECT, SUM etc);
- Stored in memory (RAM). In comparision, SQL tables are stored on hard-disk and pulled into memory while running commands.