Question

-In google Colab

from sklearn.datasets import load_breast_cancer
allData = load_breast_cancer()
X = allData['data']
y = allData['target']

-# How many features or attributes does this machine learning problem have? Hint use allData['DESCR'] to determine that.

- #Compute the shape or the dimensions of x,  and Compute the shape or the dimensions of y?

- Import a classifier, create an instance of the classifier and fit or train he classifier using the training data.

- #compute the number of true positives, true negatives, false negatives and false negative rate

Answer 1

• allData has now all the breast_cancer data using
  • allData=load_breast_cancer()
• X=allData['data'] loads the data of features into X
• Y=allData['target'] loads the target values or class labels into y (Ex- Cancer Exists or Cancer Do not Exists)
• X contains say 'i' different data examples in rows and 'j' features for each data example in columns. So, the matrix or data vector size is 'ixj'.
• Y contains 'i' rows and only one column as target values.

-If the above are taken into consideration, then the given machine learning problem as got 'j' features.

- The shape or dimensions of X can be calculated as X.shape. The shape or dimensions of y can be calculated as y.shape.

The following statements can be used to print-

• print ("X shape: " + str(X.shape))
• print ("y shape: " + str(y.shape))

-  Import the classifier using 'from sklearn.ensemble import RandomForestClassifier' (Ex- RandomForest is cosen as classifier)

clf = RandomForestClassifier(n_estimators=1000) //Classifier is defined and instance is created

clf.fit(X, y) // Classifier is fitted

- Confusion Matrix gives true positives, true negatives, false negatives and false negative rates-

Confusion Matrix is printed using the following snippet code-

Note-

• 1 indicates Cancer Exists.
• 0 indicates Cancer Do not Exits.
• The Above Snippet assumed that there are only 10 Data Examples. You can extend the same to your required size of data examples.