BYOM (Bring Your Own Model)
You can bring your own trained model into Marketing AION using PKL and
SAV file types. This section provides step-by-step instructions to use
the Marketing AION web interface for
generating information of your trained ML model.
Baselining and Testing and ML Model
- Access the Marketing AION Webpage and click MLTest.
This starts the process baseline.
- If you have downloaded the ML model file and the dataset file,
copy-and-paste the files to the Marketing AION user context.Note: Marketing AION does not identify any other user context path or root context path.
Example: /home/aionuser/mountpoint/
- Provide the Model Path.
Enter the path to your trained ML model file. The model should be saved in either joblib or pickle format.
Example Model Path:
/home/aionuser/mountpoint/HCLT/data/AB-Test-2.joblib
- Provide the Data Path.
Enter the path to your data file. The data can be either preprocessed (transformed) or raw.
Example Data Path:
/home/aionuser/mountpoint/HCLT/data/stars.csv
- Select the Training Features.
Select the training features used for training the model. Ensure that the features you select match those used during the model training to align the data correctly with the model's expectations.
- Select the Target Feature.
Specify the target feature in your dataset that the model aims to predict.
- Submit the Form.
After filling in the required information, click the Generate Baseline button. The system will process the inputs and generate the baseline report.
- View and Save the Results.
The results will be displayed on the page, including metrics such as the model name, parameters, model performance, and time latency. If needed, you can save the results to a file for future reference.