Welcome
User Guide
Model Training
Training the Model
Model training tab gives the information about the Problem type, Selected algorithms, number of rows and Training Features, Training Data, Data balancing and Number of iterations.
Details of the Trained Model
After successful training, the trained model details can be viewed.

Release Notes
Installation Guide
Developer Guide
MLAC User Guide
User Guide
- Introduction
  Marketing AION (Artificial Intelligence ON) is an AI life cycle management platform for applying machine learning to real-world problems. Marketing AION encompasses the complete pipeline from raw data ingestion to a deployable machine learning model with a less-code/no-code approach.
- Model Management
- Model Training
  - Data Upload
    Data upload tab is used to load the dataset in the GUI to train the model and get insight of it.
  - Basic Configuration
    This section provides the option to select more than one algorithm based on problem type and also provides the option to discard the unnecessary features from the dataset.
  - Updating Advanced Configuration
    To update the Advanced Configuration, complete the following steps:
  - Training the Model
    Model training tab gives the information about the Problem type, Selected algorithms, number of rows and Training Features, Training Data, Data balancing and Number of iterations.
    - Details of the Trained Model
      After successful training, the trained model details can be viewed.
- Model Prediction
  Model prediction predicts the input data based on the pattern of training data.
- Trusted AI
  This feature is available for classification and regression problem types.
- Model Monitoring
  This feature is available for classification and regression problem types.
- Utility
  For Data Generator and BYOM (Bring Your Own Model), select the Utility tab.

Details of the Trained Model

After successful training, the trained model details can be viewed.

Trained Model Info

It gives an information about- Training status, Deployment location of a model, problem type, best model (in case multiple algorithms is selected), Score of a best model, feature used, and the reports can be downloaded by clicking on Download Reports tab.

Evaluated Models

Evaluated models gives an information about the model performance based on testing score, Confidence Score and Feature Engineering Method. For classification and regression performance of seclected model is compared with the basic model such as logistic regression and linear regression respectively.

Example: When user selects Extreme Gradient Boosting (XGBoost) model for classification, its performance will be compared with Logistic regression.

Training Data Evaluation

It gives an information about training data.

Test Data Evaluation

It gives an information about test data.

Performance

Performance tab shows the graph of True Positive Rate vs False Positive Rate and Precision vs Recall, giving an idea about the model performance.

Performance Leaderboard

Performance leaderboard gives an idea about the best performing algorithm when multiple algorithms are selected based on its best score. It supports only machine learning algorithms for classification and regression problem types.

To view the leaderboard, click on the Burger icon from the right corner of the model training window. The Leaderboard appers.