search

kdmalc / fl-gestures

Develop a personalized sequential federated learning algorithm for real-time gesture classification
1 stars 0 forks source link

Score embedding clusters wrt user labels and then gesture labels #20

Open kdmalc opened 1 month ago

kdmalc commented 3 weeks ago

And also maybe the xlsx labels?

kdmalc commented 3 weeks ago

And also test the Autoencoder embeddings too? After you do PCA for this

kdmalc commented 3 weeks ago

https://scikit-learn.org/stable/api/sklearn.metrics.html

All of the following compare to sets of clustering labels. For the "ground_truth"\labels_true, iterate through the different xlsx labels (User ID, Gesture ID, hand used, muscles activated 1)

Adjusted Mutual Information Score, Adjusted Rand Score, Completeness Score, Contingency Matrix, Pair Confusion Matrix, Fowlkes-Mallows Score, Homogeneity, Completeness, and V-Measure Scores, Homogeneity Score, Mutual Information Score, Normalized Mutual Information Score, Rand Score