AttributeError: 'NoneType' object has no attribute 'split' using anaconda

[Selithrarion]

needed to upgrade threadpoolctl to 3.2
pip install threadpoolctl --upgrade
to fix the error AttributeError: 'NoneType' object has no attribute 'split' in topic 3 on the line
knn_pred = knn.predict(X_holdout_scaled)
also it was in some previous topic but dont remember where

used anaconda so im suggesting to add a hint or smth to help other people if they face it

[Selithrarion]

https://mlcourse.ai/book/topic03/assignment03_decision_trees_solution.html
line

plot_tree(
    dt, feature_names=df_train.columns, filled=True, class_names=["Won't go", "Will go"]
);

did throw an error cuz i provided an Index instead of a list so i was able to fix using df_train.columns.tolist()

[Selithrarion]

is it should be like this? data_test[c].fillna(data_train[c].mode()[0], inplace=True)

# fill missing data

for c in categorical_columns:
    data_train[c].fillna(data_train[c].mode()[0], inplace=True)
    data_test[c].fillna(data_train[c].mode()[0], inplace=True)

for c in numerical_columns:
    data_train[c].fillna(data_train[c].median(), inplace=True)
    data_test[c].fillna(data_train[c].median(), inplace=True)

[Selithrarion]

https://mlcourse.ai/book/topic06/topic6_feature_engineering_feature_selection.html#statistical-approaches - Statistical approaches section,
VarianceThreshold(0.7).fit_transform(x_data_generated).shape
and etc doesnt change the data shape

[Selithrarion]

https://mlcourse.ai/book/topic08/assignment08_implement_sgd_regressor.html
Linear regression and Stochastic Gradient Descent section

[Selithrarion]

https://mlcourse.ai/book/topic09/assignment09_time_series.html downloads an html file every time i refresh the page o_o

[Yorko]

Better late than never.

knn_pred = knn.predict(X_holdout_scaled)

Doesn't reproduce, I'm providing all dependencies with Poetry, those should work.

plot_tree(
dt, feature_names=df_train.columns, filled=True, class_names=["Won't go", "Will go"]
);

Don't see issues here.

is it should be like this? data_test[c].fillna(data_train[c].mode()[0], inplace=True)

Yes, filling in data in the test set with statistics calculated with the training set.

VarianceThreshold(0.7).fit_transform(x_data_generated).shape
and etc doesnt change the data shape

No, it actually throws away one feature

Linear regression and Stochastic Gradient Descent section

fixed in #768

https://mlcourse.ai/book/topic09/assignment09_time_series.html downloads an html file every time i refresh the page o_o

no longer see the issue (another one with Prophet will be fixed later #767)