Skip to main content

Training a Model

Welcome to the guide for training a model in Kranium AI! Once you have created your model and completed the data mapping process, training the model is a straightforward task. This guide will walk you through the steps to initiate the training process and monitor its progress from the model home tab.

1. Accessing the Model Home Tab​

Steps:

  1. Navigate to the Models Section:
    • From the main dashboard, click on the "Models" tab to access the Models section.
  2. Select Your Model:
    • In the Models listing screen, locate the model you have created and click on its name to open the model's home tab.

2. Initiating Model Training​

Steps:

  1. Locate the Start Training Button:
    • Within the model home tab, find the "Start Training" button. This button is typically prominently displayed to make it easy to start the training process.
  2. Click Start Training:
    • Click on the "Start Training" button to initiate the training process. This will trigger Kranium AI to begin training your model using the dataset and mappings you have configured.

3. Monitoring Training Progress​

Steps:

  1. View Training Progress:
    • Once training has started, the model home tab will display real-time updates on the training process. You can monitor the progress through various indicators and visualizations.
  2. Training Metrics:
    • The model home tab will show key training metrics such as loss/error rates, accuracy, precision, recall, and other relevant metrics depending on the model type. These metrics help you understand how well the model is learning from the data.
  3. Progress Bars:
    • Progress bars provide a visual representation of the training process, indicating how much of the training has been completed and the estimated time remaining.

4. Post-Training Evaluation​

Steps:

  1. Review Final Metrics:
    • Once training is complete, review the final metrics presented in the model home tab. These metrics provide a comprehensive overview of the model's performance on the training dataset.
  2. Validation and Testing:
    • Kranium AI will automatically evaluate the model on the testing dataset you specified during the configuration phase. Review the validation metrics to assess how well the model generalizes to new, unseen data.
  3. Detailed Performance Analysis: Access detailed performance reports, including confusion matrices, ROC curves, and other diagnostic tools, to thoroughly evaluate the model's effectiveness.

5. Next Steps​

Steps:

  1. Fine-Tuning:
    • If the model's performance is not satisfactory, consider fine-tuning hyperparameters from the Settings tab or making adjustments to the dataset and retraining the model.
  2. Deploying the Model:
    • If you are satisfied with the model's performance, proceed to deploy the model. Refer to the guide on deploying models in Kranium AI for detailed instructions.
  3. Model Maintenance:
    • Monitor the deployed model's performance over time and periodically retrain it with new data to ensure it continues to perform well.

Training a model in Kranium AI is designed to be an intuitive and efficient process. By following this guide, you can easily initiate and monitor the training of your model from the model home tab. With the training complete, you can evaluate the model's performance and proceed with deployment or further optimization. For any additional support or advanced configurations, refer to our support resources or contact the Kranium AI support team.