While I haven't used the sentenceEmbedding, I did try the Transfer learning algorithm with dynamic embedding of extracted features in macOS Catalina and Xcode 11, specifically CreateML.
While my 10k training data used to be consumed in less than a minute and would result in a 88Kb mlmodel file, it took an hour to train and the resulting size was now 1.5MB. Using the model to analyze 10k Rotten Tomatoes reviews used to be only as slow as I/O. Now, the analysis takes 4 minutes—whether on an iPad Pro 12.9" running iPadOS or just a simulator, no difference!
The accuracy is now an almost incredible 98%, so there is that!
databoy
Users receive reputation points by contributing quality content to the forums.
Posts include post and replies published on the forums.
Post authors can mark replies as Accepted to indicate successful solutions.
Apple Developer Forums admins can mark replies as Apple Recommended to indicate an approved solution
Last seen
User for
From
Denver, CO