Question

Answer the following question by providing Python code: In [3]: # We will use simulated data. # 3 clusters in 8 dimensional space. # Hard to visualize! X, label = make_blobs(n_samples=150, n_features=8, centers=3, cluster_std = 2, random_state=123) X.shape Out[3]: (150, 8) 1). Define a function that calculates the "total internal sum of squares". \begin{itemize} \item X: Data. \item centers: Array that contains the centroid coordinates. Calculated by kmeans. \item clusters: Array of cluster labels. Calculated by kmeans. \end{itemize} HINT: Implement the formula from the lecture note. In []: def total_internal_ss(X, centers, clusters): # <Your code goes in here> 2). Visualize as a Scatter plot the "total internal sum of squares" vs N# of clusters. Then, determine the optimal number of clusters. Is this number the same as expected?

          Answer the following question by providing Python code:
In [3]: # We will use simulated data.
# 3 clusters in 8 dimensional space.
# Hard to visualize!
X, label = make_blobs(n_samples=150, n_features=8, centers=3, cluster_std = 2, random_state=123)
X.shape
Out[3]: (150, 8)
1). Define a function that calculates the "total internal sum of squares".
\begin{itemize}
    \item X: Data.
    \item centers: Array that contains the centroid coordinates. Calculated by kmeans.
    \item clusters: Array of cluster labels. Calculated by kmeans.
\end{itemize}
HINT: Implement the formula from the lecture note.
In []: def total_internal_ss(X, centers, clusters):
# <Your code goes in here>
2). Visualize as a Scatter plot the "total internal sum of squares" vs N# of clusters.
Then, determine the optimal number of clusters.
Is this number the same as expected?

Answer the following question by providing Python code:
In [3]: # We will use simulated data.
# 3 clusters in 8 dimensional space.
# Hard to visualize!
X, label = makeblobs(nsamples=150, nfeatures=8, centers=3, clusterstd = 2, randomstate=123)
X.shape
Out[3]: (150, 8)
1). Define a function that calculates the "total internal sum of squares".

* X: Data.

* centers: Array that contains the centroid coordinates. Calculated by kmeans.

* clusters: Array of cluster labels. Calculated by kmeans.

HINT: Implement the formula from the lecture note.
In []: def totalinternalss(X, centers, clusters):
# <Your code goes in here>
2). Visualize as a Scatter plot the "total internal sum of squares" vs N# of clusters.
Then, determine the optimal number of clusters.
Is this number the same as expected?

Added by Brandi A.

Question

Please give Ace some feedback