NLU API Dashboard
1.1.0
1.1.0
  • 👋Welcome to NLU API Dashboard
  • Overview
    • 💡What You Can Do
    • ✨APIs
    • 🏄Navigation
  • Getting Started
    • 🐥Sign Up & Log In
    • 🔮Data Integration
  • Data Management
    • ⬆️Upload Data
      • Data Integration
    • 🪄Set Data
    • 🔎Filter & Search
    • ⬇️Download Data
    • 📑Named Entity Dictionary
    • 🛐Entities Settings
    • 🌲Tree Classification Category Settings
  • Model Training
    • 🚦Status
    • ⛓️Train
    • 👩‍💻Evaluate & Tune
  • Model Management
    • 📈Performance Metrics
      • Review Analysis
      • NER
      • Classification
    • 🚀Manage Model
  • 📊Analytics
  • 🔤Terms
  • ⚙️Settings
  • 😎My Account
Powered by GitBook
On this page
  • Train Model
  • Create Model
  • Deploy Model
  • Edit/Delete Model
  • Check Performance

Was this helpful?

  1. Model Management

Manage Model

Get to know how to manage your model

PreviousClassificationNextAnalytics

Last updated 2 years ago

Was this helpful?

Through the History menu in GNB, you can manage your model.

Train Model

If you press the 'Train Model' button at top of the Dashboard page, the model version is updated through training data and a new one is created.

Create Model

Once a model training is done, you can check the created model along with the following information at the top of the model list on History page.

  • Model name

  • Number of training/test data samples

  • Performance Metrics (Precision, Recall, Accuracy, F1 score)

  • Created time

Deploy Model

If you confirmed that the model has the desired performance, it can be deployed by clicking the 'Deploy' button in the Actions column.

When the deployment is complete, the button is displayed as Deployed as its status.

Edit/Delete Model

You can edit or delete the model name through the Edit/Delete menu in the right column whenever you want.

Please be careful as once deleted models cannot be restored.

Check Performance

You can also check the model's performance from History page.

You cannot start training another model on Dashboard before you finish applying the recently created model. If you want to train a new model, you can do so by deleting 'the recently created but not deployed' model.

🚀
📈Performance Metrics
GNB > Dashboard > Train Model
GNB > History > Model History > Created Model after finishing training
GNB > History > Model History > Deploy button
GNB > History > Model History > Edit/Delete Model