Resources
Resources represent the fundamental components required to create your machine learning models. They define such things as which database Telepath should read the training data from, which data should be read, and what type of model you'd like to create.
Telepath has the following resource types:
- Source - represents a database containing training data
- Pipeline - represents the query used to extract data from a Source
- ModelSpec - defines what the ML model should predict
A Resource is defined in code using the Telepath SDK by constructing an instance of its class:
// Example only. There is no actual Resource called "Resource"
new Resource(name, args, options);
All Resources have a required name argument, which must be unique across all Resources in the Project. This is a logical name that's used as a unique identifier within the scope of the Project, but it doesn't currently have any impact outside of that.
Resource names cannot be changed without recreating the entire resource. Since it's the identifier, if the name is changed, Telepath doesn't know if you're trying to create a new Resource with a new name or update the properties of an existing Resource.
That being said, recreating a Resource is a cheap and inconsequential action, since they only define how Telepath should operate and don't hold any actual data (i.e. a Source refers to a database but is not an actual database, and a Pipeline defines a query but does not actually store the data extracted by the query). This may change in time, but for now, changing a Resource name is not a big deal.