Data Sync & ETL
Set Up Medoids for Crop Anomaly Detection
When you run it manually, it calculates a cluster distance matrix over a vector collection and picks medoids for crop anomaly detection.
What it does
## Setting Up Medoids for Anomaly Detection ### Preparatory workflow to set cluster centres and cluster threshold scores, so anomalies can be detected based on these thresholds Here, we're using two approaches to set up these centres: the upper branch is the *"distance matrix approach"*, and the lower is the *"multimodal embedding model approach"*.
The steps
- When clicking ‘Test workflow’
- Total Points in Collection
- Cluster Distance Matrix
- Scipy Sparse Matrix
- Set medoid id
- Get Medoid Vector
- Prepare for Searching Threshold
- Searching Score
- Threshold Score
- Set medoid threshold score
- Split Out1
- Merge
- Textual (visual) crop descriptions
- Embed text
135 steps in total.
Similar automations