Classifier Console Guide

This page provides a guide to the Classifier entity in the Namirasoft Inference Console. It defines the concepts and configuration fields used when creating and managing Classifiers. Use this guide to understand the purpose and behavior of each setting available during Classifier configuration.

What Is a Classifier?

A Classifier reads a prompt and assigns it to the classes you define. You describe each class in plain language, and an inferencer you choose decides which classes apply. Conditions and Routers can then react to the meaning of a request rather than its exact words.

The Challenge a Classifier Solves

Word lists reach their limit fast. People phrase the same intent in endless ways, and matching words is not understanding meaning.

Common challenges include:

• Keyword fragility: A list of trigger words misses paraphrases and fires on false positives. It grows forever and is never complete.
• Meaning needs a model: Recognizing what a request is about takes AI itself. Wiring an extra AI call into routing logic is more custom code in every application.
• Labels that cannot be reused: Ad hoc intent detection lives where it was written. Conditions and routes elsewhere cannot react to it.

How Namirasoft Inference Solves the Problem

You define your classes, each with a name and a description that tells the Classifier when to select it, and you choose the inferencer that does the deciding. When a request arrives, the Classifier sends it to that inferencer together with your class list, and the assigned classes become available to your Conditions.

For example, a Classifier named “Question Type” with the classes “Coding” and “General” recognizes a code question however it is phrased, so a Router can route it to your code model.

Overview of Classifier Fields and Options

Below is a detailed explanation of the fields available when creating or managing a Classifier. Understanding these fields helps ensure your Classifier is configured correctly for your requirements.

• ID (String): This is a unique identifier automatically assigned to the Classifier when it is created. The system uses it to track, reference, and manage this specific Classifier. This value is auto-generated and cannot be modified.

• User ID (Namirasoft Account’s ID): This is the unique identifier of the Namirasoft Account user who owns this Classifier. It is used internally for permission control, audit logging, and access management.

• Workspace ID (Namirasoft Workspace’s ID): This is the identifier of the workspace this Classifier belongs to, as defined in Namirasoft Workspace. A workspace is a shared organizational space where teams group their work, projects, and members.

• Name (String): This is a label used to identify this Classifier in the console. A good name states what it sorts. For example: “Task Type Classifier”.

• Inferencer (Inferencer Type and ID): This is the inferencer that performs the classification. Any inferencer can be the deciding engine, so you can classify with a fast model to keep decisions cheap, or with a strong one when accuracy matters most.

• Multiple (Boolean): This decides how many classes a request can receive. When it is off, the Classifier assigns exactly one class. When it is on, a request can receive several classes at once.

• Classes (List): These are the classes the Classifier can assign. Each class has a Name and a Description, and the description tells the Classifier when to select that class. Write descriptions in plain language, the same way you would explain the category to a teammate.

• Additional Prompt (String): This is optional extra guidance added to the classification request. Use it to sharpen edge cases, for example: “When a request mixes topics, prefer the technical class.”

• Memory (Memory’s ID): This determines whether the Classifier keeps conversation context across requests. When a Memory is connected, relevant conversation history is stored and provided as context when the Classifier handles later requests.

• Wiper (Wiper’s ID): This determines whether stored data associated with the Classifier is cleaned up according to a defined policy. A Wiper can remove data based on time or count limits.

• Log Group (Log Group’s ID): This determines where the Classifier’s execution logs are recorded. When a Log Group, defined in Namirasoft Log, is connected, request activity and errors are recorded there for debugging and review.

• Description (String): This is optional explanatory text about this Classifier. Use it to describe what it sorts and why.

• Created At (DateTime): This is the date and time when this Classifier was created. This value is automatically generated and cannot be modified.

• Updated At (DateTime): This is the date and time when this Classifier configuration was last modified. This value is updated automatically whenever any field is changed.