Clarifai - The World's AI

Update a Clusterer Model

Step 1 - Create Model ID

Model Id

A memorable, human-readable model-id with a maximum of 48-characters

Step 2 - Set Up Model

Base Embed Model *

This is the base model to use for embeddings. This has to be one of the embed models in the app workflow. This allows you to specify the specific model in case your default workflow of your app has multiple embedding models present in it.

Base Model

Beta

Deprecated, please use eval_info.params.beta.

Coarse Clusters

Each embedding vector is first split into a fixed amount of subgroups. This is the integer value k, in k-means clustering, used to determine the numbers of centroids each subgroup is split and clustered into.

Eval Holdout Fraction

Percentage of all examples to hold out for evaluation when training.

Max Num Query Embeddings

Deprecated, please use eval_info.params.max_num_query_embeddings.

Max Visited

Deprecated, please use eval_info.params.max_visited.

Num Results Per Query

Deprecated, please use eval_info.params.num_results_per_query.

Query Holdout Fraction

Deprecated, please use eval_info.params.query_holdout_fraction.

Quota

Deprecated, please use eval_info.params.quota.

To Be Indexed Queries Fraction

Deprecated, please use eval_info.params.to_be_indexed_queries_fraction.