The NipalsPLS And NipalsPCA Classes Do Not Show "fitted" When Displayed In A Jupyter Notebook Or Ipykernel.

May 22, 2025 by ADMIN 108 views

The NipalsPLS and NipalsPCA classes do not show "fitted" when displayed in a Jupyter Notebook or IPyKernel

Introduction

When working with dimensionality reduction techniques such as Partial Least Squares (PLS) and Principal Component Analysis (PCA) in a Jupyter Notebook or IPyKernel, it can be frustrating to see that the models do not display as "fitted" even after executing the .fit method. This issue is particularly puzzling since the models are indeed fitted and ready for use. In this article, we will delve into the root cause of this problem and explore a potential solution.

Understanding the Issue

The issue at hand is related to the sklearn.utils.validation.check_is_fitted() method, which is responsible for checking whether a model has been fitted or not. This method uses two criteria to determine if a model is fitted:

Attribute checking: It checks for an attribute ending in _ but not starting in __. This is a common convention in scikit-learn to indicate that an attribute is internal and should not be accessed directly.
Callable method checking: It checks for a callable method __sklearn_is_fitted__. This method is intended to be implemented by the model's author to provide a custom way of checking if the model is fitted.

Upon further investigation, it appears that the NipalsPLS and NipalsPCA classes do not implement the __sklearn_is_fitted__ method, which is why they do not display as "fitted" in a Jupyter Notebook or IPyKernel.

Implementing the `__sklearn_is_fitted__` Method

To resolve this issue, we can implement the __sklearn_is_fitted__ method in the NipalsPLS and NipalsPCA classes. This method should return True if the model is fitted and False otherwise.

Here is an example of how we can implement this method in the NipalsPLS class:

class NipalsPLS:
    def __init__(self, n_components):
        self.n_components = n_components
        self.components_ = None
        self.scores_ = None
        self.transformed_data_ = None

    def fit(self, X):
        # implementation of the fit method
        pass

    def __sklearn_is_fitted__(self):
        return self.components_ is not None

Similarly, we can implement this method in the NipalsPCA class:

class NipalsPCA:
    def __init__(self, n_components):
        self.n_components = n_components
        self.components_ = None
        self.scores_ = None
        self.transformed_data_ = None

    def fit(self, X):
        # implementation of the fit method
        pass

    def __sklearn_is_fitted__(self):
        return self.components_ is not None

By implementing the __sklearn_is_fitted__ method, we can ensure that the NipalsPLS and NipalsPCA classes display as "fitted" in a Jupyter Notebook or IPyKernel even after executing the .fit method.

Conclusion

In conclusion, the issue of the NipalsPLS andNipalsPCAclasses not displaying as "fitted" in a Jupyter Notebook or IPyKernel is related to the lack of implementation of thesklearn_is_fittedmethod. By implementing this method, we can resolve this issue and ensure that these models display as "fitted" even after executing the.fit` method.

Example Use Case

Here is an example of how we can use the NipalsPLS and NipalsPCA classes with the __sklearn_is_fitted__ method implemented:

from nipals import NipalsPLS, NipalsPCA
import numpy as np

# create a sample dataset
X = np.random.rand(100, 10)

# create a NipalsPLS object
pls = NipalsPLS(n_components=2)

# fit the model
pls.fit(X)

# check if the model is fitted
print(pls.__sklearn_is_fitted__())  # output: True

# create a NipalsPCA object
pca = NipalsPCA(n_components=2)

# fit the model
pca.fit(X)

# check if the model is fitted
print(pca.__sklearn_is_fitted__())  # output: True

In this example, we create a sample dataset X and then create a NipalsPLS and a NipalsPCA object. We fit the models using the .fit method and then check if the models are fitted using the __sklearn_is_fitted__ method. The output of the __sklearn_is_fitted__ method is True, indicating that the models are indeed fitted.
The NipalsPLS and NipalsPCA classes do not show "fitted" when displayed in a Jupyter Notebook or IPyKernel: Q&A

Q: What is the issue with the NipalsPLS and NipalsPCA classes not displaying as "fitted" in a Jupyter Notebook or IPyKernel?

A: The issue is related to the lack of implementation of the __sklearn_is_fitted__ method in the NipalsPLS and NipalsPCA classes. This method is used by scikit-learn to check if a model is fitted or not.

Q: Why is the `__sklearn_is_fitted__` method important?

A: The __sklearn_is_fitted__ method is important because it allows scikit-learn to check if a model is fitted or not. If a model is not fitted, it cannot be used for predictions or other tasks. By implementing this method, we can ensure that the NipalsPLS and NipalsPCA classes display as "fitted" in a Jupyter Notebook or IPyKernel.

Q: How can I implement the `__sklearn_is_fitted__` method in the `NipalsPLS` and `NipalsPCA` classes?

A: To implement the __sklearn_is_fitted__ method, you need to add a method to the NipalsPLS and NipalsPCA classes that returns True if the model is fitted and False otherwise. Here is an example of how you can implement this method:

class NipalsPLS:
    def __init__(self, n_components):
        self.n_components = n_components
        self.components_ = None
        self.scores_ = None
        self.transformed_data_ = None

    def fit(self, X):
        # implementation of the fit method
        pass

    def __sklearn_is_fitted__(self):
        return self.components_ is not None

Similarly, you can implement this method in the NipalsPCA class:

class NipalsPCA:
    def __init__(self, n_components):
        self.n_components = n_components
        self.components_ = None
        self.scores_ = None
        self.transformed_data_ = None

    def fit(self, X):
        # implementation of the fit method
        pass

    def __sklearn_is_fitted__(self):
        return self.components_ is not None

Q: What are the benefits of implementing the `__sklearn_is_fitted__` method?

A: The benefits of implementing the __sklearn_is_fitted__ method include:

Ensuring that the NipalsPLS and NipalsPCA classes display as "fitted" in a Jupyter Notebook or IPyKernel.
Allowing scikit-learn to check if a model is fitted or not.
Enabling the use of the NipalsPLS and NipalsPCA classes for predictions and other tasks.

Q: How can I use the `NipalsPLS` and `NipalsPCA` classes with the `__sklearn_is_fitted__` method implemented?

A: To use the NipalsPLS and NipalsPCA classes with the __sklearn_is_fitted__ method implemented, you can follow these steps:

Create a sample dataset.
Create a NipalsPLS or NipalsPCA object.
Fit the model using the .fit method.
Check if the model is fitted using the __sklearn_is_fitted__ method.

Here is an example of how you can use the NipalsPLS and NipalsPCA classes:

from nipals import NipalsPLS, NipalsPCA
import numpy as np

# create a sample dataset
X = np.random.rand(100, 10)

# create a NipalsPLS object
pls = NipalsPLS(n_components=2)

# fit the model
pls.fit(X)

# check if the model is fitted
print(pls.__sklearn_is_fitted__())  # output: True

# create a NipalsPCA object
pca = NipalsPCA(n_components=2)

# fit the model
pca.fit(X)

# check if the model is fitted
print(pca.__sklearn_is_fitted__())  # output: True