Introduction

Upon creating a new project, AryaXAI automatically trains a default AutoML model, 'XGBoost_default', for default prediction and explainability. However, users have the option to utilize their own model for explainability through:

1. Uploading own model.
2. Training a model using AryaXAI's built-in modelling techniques, which include:

• XGBoost
• LGBoost
• CatBoost
• RandomForest
• SGD (Stochastic Gradient Descent)
• Logistic Regression
• Linear Regression
• GaussianNaiveBayes

Users can fine-tune these models and adjust the hyperparameters according to their requirements.

Uploading own model

Before uploading a model, ensure that the corresponding features have already been uploaded through the data upload process.

Select the tag in which your model is trained on. Also, you can define the test data when you are uploading the model

project.upload_model(model_path='/content/xgb_sample_model.pkl',
                     model_name = 'XgbCustom',
                     model_data_tags = ['Training'],
                     model_type='Xgboost',
                     model_architecture='machine_learning')

‍

Help on method upload_model in module aryaxai.core.project

help(project.upload_model)

‍

Supported models

Following are the models that you can upload to AryaXAI:

NOTE: Deep Learning is not available at the moment. In the meantime, you can use ML.

‍

project.upload_model_types()

{'model_architecture': {'machine_learning': ['Xgboost',
   'Lgboost',
   'CatBoost',
   'Random_forest',
   'Linear_Regression',
   'Logistic_Regression',
   'Gaussian_NaiveBayes',
   'SGD'],
  'deep_learning': ['tensorflow', 'pytorch']},
 'prediction_type': ['classification', 'regression']}

‍

NOTE: If you encounter an error, please ensure that the columns in the files you uploaded match those used in your model. We use your model settings to validate that the data in these tags is consistent. The framework will only proceed with adding the model if all features are available.

‍

To view uploaded model info

project.models()

‍

Delete uploaded file

project.delete_file('file name')

‍