davrot/pytutorial
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update README.md

Browse source 
Signed-off-by: David Rotermund <54365609+davrot@users.noreply.github.com>
This commit is contained in:
David Rotermund 2023-12-19 14:37:55 +01:00 committed by GitHub
parent f99563f1c8
commit c74c3e120c
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 108 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

109
scikit-learn/kmeans/README.md
View file

@ -72,7 +72,7 @@ data_b = np.concatenate((b_x, b_y), axis=1)
data = np.concatenate((data_a, data_b), axis=0)
kmeans = KMeans(n_clusters=2)
kmeans = KMeans(n_clusters=2, n_init = 10)
kmeans.fit(data)
@ -90,4 +90,111 @@ plt.show()
![image1](image1.png)
> **labels_** : ndarray of shape (n_samples,)
> Labels of each point
## What does the algorithm „think“ where the data points belong?​
```python
import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
rng = np.random.default_rng(1)
a_x = rng.normal(1.5, 1.0, size=(1000))[:, np.newaxis]
a_y = rng.normal(3.0, 1.0, size=(1000))[:, np.newaxis]
data_a = np.concatenate((a_x, a_y), axis=1)
b_x = rng.normal(0.0, 1.0, size=(1000))[:, np.newaxis]
b_y = rng.normal(0.0, 1.0, size=(1000))[:, np.newaxis]
data_b = np.concatenate((b_x, b_y), axis=1)
data = np.concatenate((data_a, data_b), axis=0)
kmeans = KMeans(n_clusters=2, n_init = 10)
kmeans.fit(data)
labels = kmeans.labels_
idx_0 = np.where(labels == 0)[0]
idx_1 = np.where(labels == 1)[0]
plt.plot(data[idx_0, 0], data[idx_0, 1], "r.")
plt.plot(data[idx_1, 0], data[idx_1, 1], "b.")
plt.plot(
kmeans.cluster_centers_[0, 0], kmeans.cluster_centers_[0, 1], "k*", markersize=12
)
plt.plot(
kmeans.cluster_centers_[1, 0], kmeans.cluster_centers_[1, 1], "k*", markersize=12
)
plt.show()
```
![image2](image2.png)
## [predict](https://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans.predict)
```python
predict(X, sample_weight='deprecated')
```
> Predict the closest cluster each sample in X belongs to.
>
> In the vector quantization literature, cluster\_centers\_ is called the code book and each value returned by predict is the index of the closest code in the code book.
```python
import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
rng = np.random.default_rng(1)
a_x = rng.normal(1.5, 1.0, size=(1000))[:, np.newaxis]
a_y = rng.normal(3.0, 1.0, size=(1000))[:, np.newaxis]
data_a = np.concatenate((a_x, a_y), axis=1)
b_x = rng.normal(0.0, 1.0, size=(1000))[:, np.newaxis]
b_y = rng.normal(0.0, 1.0, size=(1000))[:, np.newaxis]
data_b = np.concatenate((b_x, b_y), axis=1)
data = np.concatenate((data_a, data_b), axis=0)
kmeans = KMeans(n_clusters=2, n_init=10)
kmeans.fit(data)
x = np.linspace(data[:, 0].min(), data[:, 0].max(), 100)
y = np.linspace(data[:, 1].min(), data[:, 1].max(), 100)
xx, yy = np.meshgrid(x, y)
xx_r = xx.ravel()[:, np.newaxis]
yy_r = yy.ravel()[:, np.newaxis]
print(xx.shape) # -> (100, 100)
print(xx_r.shape) # -> (10000, 1)
print(yy.shape) # -> (100, 100)
print(yy_r.shape) # -> (10000, 1)
coordinates = np.concatenate((xx_r, yy_r), axis=1)
print(coordinates.shape) # -> (10000, 2)
labels = kmeans.predict(coordinates)
idx_0 = np.where(labels == 0)[0]
idx_1 = np.where(labels == 1)[0]
plt.plot(coordinates[idx_0, 0], coordinates[idx_0, 1], "r.")
plt.plot(coordinates[idx_1, 0], coordinates[idx_1, 1], "b.")
plt.plot(
kmeans.cluster_centers_[0, 0], kmeans.cluster_centers_[0, 1], "k*", markersize=12
)
plt.plot(
kmeans.cluster_centers_[1, 0], kmeans.cluster_centers_[1, 1], "k*", markersize=12
)
plt.show()
```
![image3](image3.png)