# This is Jeopardy Challenge Project (Python, Pandas)

Hello, here : GitHub - HarryTutle/jeopardy-game: a game with jeopardy you will find my little version of the jeopardy game.

Hello, this is my notebook for the Jeopardy Project:

I use the notebooks to explore the data and also to take notes about what I am learning/trying as I go through the project… so it is a bit long. There are also links at the bottom with all the references that I used.

Hope it helps someone. Feedback is welcome!

Note: The rendering of notebooks in GitHub is not that great (and sometimes it just doesn’t work). nbviewer, from the Jupyter project, works better. Just copy/paste the URL from GitHub there.

It’s about calculating the mean for the question values, but the final round questions actually don’t have a (\$) value and so the solution code recommends setting all those (it’s 3654 of them) to zero:

``````jeopardy_data["Float Value"] = jeopardy_data["Value"].apply(
lambda x: float(x[1:].replace(",", "")) if x != "None" else 0
)
``````

This results in an average value of 740 (rounded). Wouldn’t it be more sound to ignore those questions alltogether? Is it really fair to have them lowering the mean of all the other questions?

I tried to do so by setting all ‘None’ strings to actual NaN values:

``````df = pd.read_csv("jeopardy.csv", na_values="None")
``````

and converted the value column to float type inplace:

``````df.value = df.value.str[1:].str.replace(",", "").astype(float)
``````

which should make the mean() function ignore the NaNs and gives me a result of 753 (rounded).

Well, what do you think?

Hi everyone,

You can find my solution for the project here: https://github.com/Mikheltodd/myDataScience/tree/main/PythonProjects/This%20is%20Jeopardy!
https://github.com/Mikheltodd/myDataScience.git

I created a function that includes the name of the column as a parameter and another to specify how to apply the filter (using all or any functions).

Best regards to all,