Introduction to Supervised Learning with scikit-learn

In this lab, we will cover **Evaluation metrics** for Classification. There are many ways to measure classification performance: Accuracy, and confusion matrices, are some of the most popular metrics. Precision & recall are widely used metrics for classification problems.

All our Data Science projects include bite-sized activities to test your knowledge and practice in an environment with constant feedback.

All our activities include solutions with explanations on how they work and why we chose them.

input

A spam recognition classifier is described by the following confusion matrix:

TP, TN, FP, FN = 4, 91, 1, 4

Compute the accuracy and insert the answer in the box below.

Round to two significant decimals

input

A spam recogition classifier is described by the following confusion matrix:

TP, TN, FP, FN = 0, 95, 5, 0

Compute the accuracy and insert the answer in the box below.

Round to two significant decimals

input

Compute the precision based on the following results:

TP = 114 FP = 14

Round to two significant decimals

input

Compute the recall based on the following results:

TP = 114 FN = 0

Round to two significant decimals.

input

Compute the F1- score based on the following results:

TP, FP, FN = 2.00, 1.00, 90.00

Round to two significant decimals

codevalidated

Let's now evaluate another example of how you could calculate some evaluation metrics.

```
# True labels of the data
y_true = [0, 1, 0, 1, 1, 0, 1, 1, 0, 0]
# Predicted labels of the data
y_pred = [0, 1, 0, 1, 0, 1, 1, 0, 1, 0]
```

Store the result in the variables `f1`

,`accuracy`

,`precision`

and `recall`

.

input

```
X_train = [[4,2,1],[3,4,6],[5,6,7],[8,9,7]]
y_train = [1,2,1,2]
X_test = [[4,3,1],[2,4,3],[5,6,1],[5,9,9]]
y_test = [1,2,2,2]
```

Use random_state=0 in the model and average='weighted'to calculate the precision. round to two decimal places the result.

input

```
X_train = [[4,2,1],[3,4,6],[5,6,7],[8,9,7]]
y_train = [1,2,1,2]
X_test = [[4,3,1],[2,4,3],[5,6,1],[5,9,9]]
y_test = [1,2,2,2]
```

Use random_state=0 in the model and average='weighted'to calculate the recall. round to two decimal places the result.

multiplechoice

```
from sklearn.datasets import make_blobs
X_train, y_train = make_blobs(n_samples=100, centers=2,
random_state=0, cluster_std=2.3)
X_test, y_test = make_blobs(n_samples=10, centers=2,
random_state=0, cluster_std=4.5)
```

Use random_state=0 in the model and average='weighted'to calculate the F1-score. round to two decimal places the result.

multiplechoice

Let’s say we have a machine that classifies if a fruit is an apple or not.

This project is part of

Explore other projects