Revisions to What's the difference between finding the average Euclidean distance and using inertia_ in KMeans in sklearn?

fixing my horrific noob formatting

Source Link

edited Jan 23, 2019 at 11:23

69
4

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data,
                              ,        kmeanModel.cluster_centers_,
                              ,        'euclidean'),
                              ,  axis = 1)) / data.shape[0])

enter image description here

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

enter image description here

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data, 
                              ,        kmeanModel.cluster_centers_, 
                              ,        'euclidean'), 
                              ,  axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) / data.shape[0])

enter image description here

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

enter image description here

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data,
                                      kmeanModel.cluster_centers_,
                                      'euclidean'),
                                axis = 1)) / data.shape[0])

enter image description here

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

enter image description here

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data, 
                                      kmeanModel.cluster_centers_, 
                                      'euclidean'), 
                                axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

added 34 characters in body

Source Link

edited May 24, 2018 at 14:10

lstodd

69
4

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) / data.shape[0])

enter image description here

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

enter image description here

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

enter image description here

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) / data.shape[0])

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

enter image description here

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) / data.shape[0])

enter image description here

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

enter image description here

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

deleted 42 characters in body

Source Link

edited May 24, 2018 at 14:02

lstodd

69
4

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) / data.shape[0])

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

enter image description here

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) / data.shape[0])

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

I've found two different approaches online when using the Elbow Method to determine the optimal number of clusters for K-Means.

One approach is to use the following code:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) / data.shape[0])

Another is to use inertia_ from sklearn.cluster.KMeans:

distortions_3.append(kmeanModel.inertia_)

When I plot the results (using the same random states) both give different results but I'm not sure what the differences are, can anyone help?

Edit: If I replace the normalisation factor / data.shape[0] with squared **2 as suggested below, then I still don't get the same as for the inertia plot:

distortions_2.append(sum(np.min(cdist(data
                              , kmeanModel.cluster_centers_
                              , 'euclidean')
                              , axis = 1)) ** 2)

Using squared just makes the plot a little smoother, but definitely not the same as using intertia_, I'm just not quite sure how inertia_ is calculated and what it means.

enter image description here

added 166 characters in body

Source Link

edited May 24, 2018 at 13:40

lstodd

69
4

Loading

added 4 characters in body

Source Link

edited May 23, 2018 at 13:11

lstodd

69
4

Loading

added 550 characters in body

Source Link

edited May 23, 2018 at 12:46

lstodd

69
4

Loading

Source Link

asked May 22, 2018 at 14:21

lstodd

69
4

Loading

Stack Exchange Network

Return to Question