name 'target' is not defined Feature Selection-CodePudding

I am working on feature selection based on the LOFO Importance selection algorithm. I have defined the target value but it gives an error that the target value is not defined.

import pandas as pd
from sklearn.model_selection import KFold
from lofo import LOFOImportance, Dataset, plot_importance
%matplotlib inline

# import data
train_df = pd.read_csv("/content/heat_pipe.csv")

train_df.head()

# # extract a sample of the data
sample_df = train_df.sample(frac=0.01, random_state=0)
sample_df.sort_values("Date", inplace=True)

# # define the validation scheme
cv = KFold(n_splits=4, shuffle=True, random_state=0)

# # define the binary target and the features
dataset = Dataset(df=sample_df, target="day_of_the_week", features=[col for col in train_df.columns if col != target])

# # define the validation scheme and scorer. The default model is LightGBM
lofo_imp = LOFOImportance(dataset, cv=cv, scoring="roc_auc")

# # # get the mean and standard deviation of the importances in pandas format
importance_df = lofo_imp.get_importance()

# # # plot the means and standard deviations of the importances
plot_importance(importance_df, figsize=(12, 20))

Error

NameError                                 Traceback (most recent call last)
<ipython-input-33-cca2efb2dc97> in <module>()
     34 
     35 # # define the binary target and the features
---> 36 dataset = Dataset(df=sample_df, target="day_of_the_week", features=[col for col in train_df.columns if col != target])
     37 
     38 # # # define the validation scheme and scorer. The default model is LightGBM

<ipython-input-33-cca2efb2dc97> in <listcomp>(.0)
     34 
     35 # # define the binary target and the features
---> 36 dataset = Dataset(df=sample_df, target="day_of_the_week", features=[col for col in train_df.columns if col != target])
     37 
     38 # # # define the validation scheme and scorer. The default model is LightGBM

NameError: name 'target' is not defined

CodePudding user response：

There is a confusion with your usage of

dataset = Dataset(df=sample_df, target="day_of_the_week", features=[col for col in train_df.columns if col != target])

The target="day_of_the_week" is a kwarg (a named parameter for the Dataset() class. It isn't assigning the value of "day_of_the_week" to target.

Try adding target = "day_of_the_week" before that line:


import pandas as pd
from sklearn.model_selection import KFold
from lofo import LOFOImportance, Dataset, plot_importance
%matplotlib inline

# import data
train_df = pd.read_csv("/content/heat_pipe.csv")

train_df.head()

# # extract a sample of the data
sample_df = train_df.sample(frac=0.01, random_state=0)
sample_df.sort_values("Date", inplace=True)

# # define the validation scheme
cv = KFold(n_splits=4, shuffle=True, random_state=0)

# set target
target = "day_of_the_week"

# # define the binary target and the features
dataset = Dataset(df=sample_df, target=target, features=[col for col in train_df.columns if col != target])

# # define the validation scheme and scorer. The default model is LightGBM
lofo_imp = LOFOImportance(dataset, cv=cv, scoring="roc_auc")

# # # get the mean and standard deviation of the importances in pandas format
importance_df = lofo_imp.get_importance()

# # # plot the means and standard deviations of the importances
plot_importance(importance_df, figsize=(12, 20))