Classification API

Gainly offers multiple API endpoints that can be used for classification tasks:

• Categorize
• Classify
• Analyze Sentiment
• Predict

Select the Best Endpoint

Custom ML Models

Custom ML models refers to machine learning (ML) models that you have trained using your own historical data. You can train a custom ML model from your Gainly Dashboard under Settings > Custom Models.

Data for Training Custom ML Models

To train a custom ML model, we recommend at least 1000 samples to train on. These samples must be:

• High quality
• Balanced
• Representative of your domain
• Consistent

Use the following flowchart to help you select the best endpoint for your classification tasks.

flowchart TD
    A[Do you have historical data to train a custom ML model?] -->|Yes| B[Do you want to categorize based on text content or attributes?]
    B -->|Text| C[Classify]
    B -->|Attributes| D[Predict]
    A -->|No| E[Do you want to classify into sentiments?]
    E -->|Yes| F[Analyze Sentiment]
    E -->|No| G[Categorize]

    click C "../../api-reference/classify/" "Go to Classify API Reference"
    click D "../../api-reference/predict/" "Go to Predict API Reference"
    click F "../../api-reference/analyze-sentiment/" "Go to Analyze Sentiment API Reference"
    click G "../../api-reference/categorize-documents/" "Go to Categorize API Reference"