From the course: Using AI with Filemaker Pro
Unlock this course with a free trial
Join today to access over 24,300 courses taught by industry experts.
Working with cosine similarity - FileMaker Pro Tutorial
From the course: Using AI with Filemaker Pro
Working with cosine similarity
- [Instructor] When you perform a semantic find in your FileMaker app, you'll be provided with the return count that you indicated in your performed semantic find script step. In our case, it's 11. We discussed that the 11 records in this case are ordered by their cosine similarity, with values closer to one listed first in the order, indicating that they have higher semantic similarity than the lowest semantic similarity, in this case would be the 11th record in the record count. So the order is very meaningful. They are ordered by cosine similarity with the first record being the most similar. Now, we cannot visually see the similarity values since they are calculated behind the scenes, so why don't we create a calculation that displays those values? So logically we may think about creating a calculation field to reveal those cosine values, but there's a couple things that we might want to consider in our specific use case in the meetings file here. First of all, when you create a…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
(Locked)
AI in FileMaker Pro 20244m 32s
-
(Locked)
Setting up AI integration3m 50s
-
(Locked)
Understanding insert embeddings4m 7s
-
(Locked)
Understanding GetEmbedding2m 14s
-
(Locked)
Working with embeddings5m 10s
-
(Locked)
Understanding Perform Semantic Find4m 13s
-
(Locked)
Working with Perform Semantic Find5m 46s
-
(Locked)
Testing Perform Semantic Find4m 43s
-
(Locked)
Understanding cosine similarity3m 11s
-
(Locked)
Working with cosine similarity4m 44s
-
Working with GetEmbedding6m 4s
-
(Locked)
Understanding optimizing AI usage4m 40s
-
(Locked)
Working with Set AI Call Logging3m 8s
-
(Locked)
Working with token counts6m 50s
-
(Locked)
-