Chapter 10 typo

[rickiepark]

(p326) In a paragraph below code block, "Using the 'absolute_loss' as an argument for the residual_metric" should be "Using the 'absolute_loss' as an argument for the loss".

[rickiepark]

(p315) In code sns.pairplot(df[cols], size=2.5), size param is changed to height.

(p324) 1st paragraph is not false but readers may misunderstand that LinearRegression class use LIBLINEAR because the following code use LinearRegression class. As you know this class use scipy.linalg.lstsq function. It would be better to clarify usage of LIBLINEAR.

(p338) Maybe .f(x)=2^{-x} means f(x)=e^{-x}.

[rickiepark]

(p314) in the last line of Note box,
df = pd.read_csv('./housing.data.txt'), sep='\s+')
should be
df = pd.read_csv('./housing.data.txt', sep='\s+')

[rasbt]

Thanks again for all your useful comments. I really appreciate it, and I agree with all your points.

(p326) In a paragraph below code block, "Using the 'absolute_loss' as an argument for the residual_metric" should be "Using the 'absolute_loss' as an argument for the loss".

Oh yes, not sure why I wrote residual_metric

(p315) In code sns.pairplot(df[cols], size=2.5), size param is changed to height.

Not using seaborn in the 3rd edition anymore (in favor of just matplotlib) so that issue should go away.

(p324) 1st paragraph is not false but readers may misunderstand that LinearRegression class use LIBLINEAR because the following code use LinearRegression class. As you know this class use scipy.linalg.lstsq function. It would be better to clarify usage of LIBLINEAR.

Good point. I think I was conflating that with the Lasso and Ridge implementations

(p338) Maybe .f(x)=2^{-x} means f(x)=e^{-x}.

Absolutely!

(p314) in the last line of Note box, ...

Good catch!