I have a training data set that has categorical features on which I use pd.get_dummies to one hot encode. This produces a data set with n features. I then train a classification model on this data set with n features. If I now get some new data with the same categorical features and again perform one hot encoding, the resultant number of features is m < n.
I cannot predict the classes of the new data set if the dimensions don't match with the original training data.
Is there a way to include all of the original n features in the new data set after one hot encoding?
EDIT: I am using sklearn.ensemble.RandomForestClassifier as my classification library.
For example ,
You have tradf with column ['A_1','A_2']
With your new df you have column['A'] but only have one category 1 , you can do
pd.get_dummies(df).reindex(columns=tradf.columns,fill_value=0)
Related
How do you measure with features of your dataframe are important for your Kmeans model?
I'm working with a dataframe that has 37 columns of which 33 columns are of categorical data.
These 33 data columns go through one-hot-encoding and now I have 400 columns.
I want to see which columns have an impact on my model and which don't.
Is there a method for this or do I loop this?
For categorical values there is K-Modes and for mixed (categorical and continuous values) there is K-Prototype. That might be worth trying and potentially easier to evaluate. You wouldn't use one-hot encoding there though.
I have a dataset with more 1500+ category in single feature. how to encode these feature ? I tried with target encoding but there is category mismatch in training and test dataset. For example in training dataset there is A,B,C category of feature X while in test dataset there is A,B,D,F category of feature X ?
How to deal with category mismatch and encode categorical variable with high cardinality?
I am wondering what is the difference between pandas' get_dummies() encoding of categorical features as compared to the sklearn's OneHotEncoder().
I've seen answers that mention that get_dummies() cannot produce encoding for categories not seen in the training dataset (answers here). However, this is a result of having performed the get_dummies() separately on the testing and training datasets (which can give inconsistent shapes). On the other hand, if we applied the get_dummies() on the original dataset, before splitting it, I think the two methods should give identical results. Am I wrong? Would that cause problems?
My code is currently working like the one below:
def one_hot_encode(ds,feature):
#get DF of dummy variables
dummies = pd.get_dummies(ds[feature])
#One dummy variable to drop (Dummy Trap)
dummyDrop = dummies.columns[0]
#Create a DF from the original and the dummies' DF
#Drop the original categorical variable and the one dummy
final = pd.concat([ds,dummies], axis='columns').drop([feature,dummyDrop], axis='columns')
return final
#Get data DF
dataset = pd.read_csv("census_income_dataset.csv")
columns = dataset.columns
#Perform one-hot-encoding on the DF (See function above) on categorical features
features = ["workclass","marital_status","occupation","relationship","race","sex","native_country"]
for f in features:
dataset = one_hot_encode(dataset,f)
#Re-order to get ouput feature in last column
dataset = dataset[[c for c in dataset.columns if c!="income_level"]+["income_level"]]
dataset.head()
If you apply get_dummies() and OneHotEncoder() in the general dataset, you should obtain the same result.
If you apply get_dummies() in the general dataset, and OneHotEncoder() in the train dataset, you will probably obtain a few (very small) differences if in the test data you have a "new" category. If not, they should have the same result.
The main difference between get_dummies() and OneHotEncoder() is how they work when you are using this model in real life (or in production) and your receive a "new" class of a categorical column that you haven't faced before
Example: Imagine your category "sex" can be only: male or female, and you sold your model to a company. What will happen if now, the category "sex" receives the value: "NA" (not applicable)? (Also, you can image that "NA" is an option, but it only appear 0.001%, and casually, you don't have any of this value in your dataset)
Using get_dummies(), you will have a problem, since your model is trained for only 2 different categories of sex, and now, you have a different and new category that the model can't hand with it.
Using OneHotEncoder(), will allow you to "ignore" this new category that your model can't face, allowing you to keep the same shape between the model input, and your new sample input.
That's why people uses OneHotEncoder() in train set and not in the general dataset, they are "simulating" this type of success (having "new" class you haven't faced before in a categorical column)
I am using Lime (Local Interpretable Model-agnostic Explanations) with mixed feature types in order to evaluate my model predictions for classification task. Does anyone know how to specify binary features in lime.lime_tabular.LimeTabularExplainer() method. How actually LIME handles these types of features (more features with only 1's and 0's)?
I think your should declare your binary features as categorical features in order to allow your Lime explainer to use its sampling mechanism efficiently when performing local perturbation around the studied sample.
You can do it using the categorical_features keyword parameter in the LimeTabularExplainer constructor.
my_binary_feature_column_index = 0 # put your column index here
explainer = LimeTabularExplainer(my_data, categorical_features=[my_binary_feature_column_index], categorical_name={my_binary_feature_column_index: ["foo", "bar", "baz"]})
categorical_features is a list of categorical column indexes, and
categorical_name is a dictionary containing a map of column index and list of category names.
As it is mentionned in the LIME code :
Explains predictions on tabular (i.e. matrix) data.
For numerical features, perturb them by sampling from a Normal(0,1) and
doing the inverse operation of mean-centering and scaling, according to the
means and stds in the training data. For categorical features, perturb by
sampling according to the training distribution, and making a binary
feature that is 1 when the value is the same as the instance being
explained.
So, categorical features are one hot encoded under the hood and the value 0 or 1 is used according to the feature distribution in your training dataset (unless you chose to use a LabelEncoder, which will result in LIME processing the feature as a continuous variable).
A good tutorial is available in the LIME project: https://github.com/marcotcr/lime/blob/master/doc/notebooks/Tutorial%20-%20continuous%20and%20categorical%20features.ipynb
I have a large multi dimensional unlabelled dataset of cars (price, mileage, horsepower, ...) for which I want to find outliers. I decided to use the sklearn OneClassSVM to build a decision boundary and have two main issues with my approach:
My dataset contains a lot of missing values. Is there a way to make the svm classify the data with missing features as an inlier if any possible values for the missing features would be an inlier?
I now want to add a feedback loop of manual moderated outliers. The manually moderated data should improve the classification of the SVM. I've read about the LabelSpreading model for semi-supervised learning. Would it be feasible to feed the classification output of the OneClassSVM to the LabelSpreading model and retrain this model when a sufficient amount of records are manually validated?
For the first question. You could use an sklearn.preprocessing.imputer to impute the missing values by mean or median:
http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.Imputer.html
You could add some boolean Features that recode if some of the other features had NaNs. So if you have features X_1, X_2 you add the boolean features
X_1_was_NaN and X_2_was_NaN
that are 1 if X_1==NaN or X_2==NaN. If X is your original pd.DataFrame you can create this by
X = pd.DataFrame()
# Create your features here
# Get the locations of the NaNs
X_2 = 1.0 * X.isnull()
# Rename columns
X_2.rename(columns=lambda x: str(x)+"_has_NaN", inplace=True)
# Paste them together
X = pd.concat([X, X_2], axis=1)