📄️ GPU Models
navio supports the use of a GPU for accelerated inference.
📄️ Additional Fields
In many cases it is useful to output additional information from the model besides the actual prediction list. This can be achieved very simply by adding keys to the output dictionary of the model’s predict method.
📄️ Time Series Models
To create a time series model it is necessary to specify a single dateTime column in the model's Request Schema.
📄️ Image Models
One way to send image data to a navio prediction endpoint is to encode it as a base64 string. This is possible with any popular python image processing library, such as PIL:
📄️ Custom Explanations
Custom explanations are enabled if a model is assigned a data set and the MLmodel YAML file specifies that explanations are provided by the model using Plotly:
📄️ Model Retraining
A navio MLflow model can be retrained by associating a Retrainer with it. A Retrainer is itself an MLflow model which implements the training and packaging steps for the retrained model in its predict method. During retraining, the model_input frame passed to the Retrainer's predict() method will contain one row with the following fields: