Question 6 - pipeline

M1. The Predictive Modeling Pipeline M1. Wrap-up quiz 1
#1

Hi,

I have difficulties in coding a pipeline that can process both the numerical and categorical features together.

I first created two dataframes : data_numerical and data_category.

Then I tried this :

make_column_transformer(
     (StandardScaler(), data_numerical),
     (OneHotEncoder(handle_unknown="ignore"), data_category))

model2 = make_pipeline(
    make_column_transformer, LogisticRegression(max_iter=500))

data2 = data_numerical + data_category

cv_results2 = cross_validate(model2, data2, target, cv=10)
cv_results2

It doesn’t work. Could you help me ?
Thanks

#2

Can you provide the traceback error that you obtain?

This line is suspicious:

data2 = data_numerical + data_category

Here I assume that we concatenate the column names. In cross_validate then you are passing column names instead of the data directly.

#3

Thanks for your answer.
I tried this :

from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.compose import make_column_transformer

make_column_transformer(
     (StandardScaler(), data[numerical_features]),
     (OneHotEncoder(handle_unknown="ignore"), data.drop(columns=numerical_features)))

model2 = make_pipeline(
    make_column_transformer, LogisticRegression(max_iter=500))

cv_results2 = cross_validate(model2, data, target, cv=10)
cv_results2

Here is the traceback error :

ValueError                                Traceback (most recent call last)
Cell In[12], line 4
      1 model2 = make_pipeline(
      2     make_column_transformer, LogisticRegression(max_iter=500))
----> 4 cv_results2 = cross_validate(model2, data, target, cv=10)
      5 cv_results2

File /opt/conda/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:285, in cross_validate(estimator, X, y, groups, scoring, cv, n_jobs, verbose, fit_params, pre_dispatch, return_train_score, return_estimator, error_score)
    265 parallel = Parallel(n_jobs=n_jobs, verbose=verbose, pre_dispatch=pre_dispatch)
    266 results = parallel(
    267     delayed(_fit_and_score)(
    268         clone(estimator),
   (...)
    282     for train, test in cv.split(X, y, groups)
    283 )
--> 285 _warn_or_raise_about_fit_failures(results, error_score)
    287 # For callabe scoring, the return type is only know after calling. If the
    288 # return type is a dictionary, the error scores can now be inserted with
    289 # the correct key.
    290 if callable(scoring):

File /opt/conda/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:367, in _warn_or_raise_about_fit_failures(results, error_score)
    360 if num_failed_fits == num_fits:
    361     all_fits_failed_message = (
    362         f"\nAll the {num_fits} fits failed.\n"
    363         "It is very likely that your model is misconfigured.\n"
    364         "You can try to debug the error by setting error_score='raise'.\n\n"
    365         f"Below are more details about the failures:\n{fit_errors_summary}"
    366     )
--> 367     raise ValueError(all_fits_failed_message)
    369 else:
    370     some_fits_failed_message = (
    371         f"\n{num_failed_fits} fits failed out of a total of {num_fits}.\n"
    372         "The score on these train-test partitions for these parameters"
   (...)
    376         f"Below are more details about the failures:\n{fit_errors_summary}"
    377     )

ValueError: 
All the 10 fits failed.
It is very likely that your model is misconfigured.
You can try to debug the error by setting error_score='raise'.

Below are more details about the failures:
--------------------------------------------------------------------------------
10 fits failed with the following error:
Traceback (most recent call last):
  File "/opt/conda/lib/python3.10/site-packages/sklearn/model_selection/_validation.py", line 686, in _fit_and_score
    estimator.fit(X_train, y_train, **fit_params)
  File "/opt/conda/lib/python3.10/site-packages/sklearn/pipeline.py", line 402, in fit
    Xt = self._fit(X, y, **fit_params_steps)
  File "/opt/conda/lib/python3.10/site-packages/sklearn/pipeline.py", line 340, in _fit
    self._validate_steps()
  File "/opt/conda/lib/python3.10/site-packages/sklearn/pipeline.py", line 231, in _validate_steps
    raise TypeError(
TypeError: All intermediate steps should be transformers and implement fit and transform or be the string 'passthrough' '<function make_column_transformer at 0x7f3d217ac700>' (type <class 'function'>) doesn't
#4

You need to pass a ColumnTransformer object to make_pipeline something like:


preprocessor = make_column_transformer(...)

model2 = make_pipeline(preprocessor, LogisticRegression(max_iter=500)))

In your snippet you pass make_column_transformer and the error kind of mentions that this is incorrect:

TypeError: All intermediate steps should be transformers and implement fit and transform or be the string 'passthrough' '<function make_column_transformer at 0x7f3d217ac700>' (type <class 'function'>) doesn't

Some advice when you are debugging this kind of error of error in cross_validate and all your fits failed, it helps a bit to do model2.fit(data, target) you will directly get the single error at the bottom of your traceback.

Also I edited your previous message to make it more readable with markdown (in particular using triple backquotes).

#5

Thanks