Crate rten_generate

Crate rten_generate 

Source
Expand description

Utilities to simplify running auto-regressive RTen models such as transformer decoders.

For working examples, see the examples in the rten-examples crate which import rten_generate.

Re-exports§

pub use generator::Generator;
pub use generator::GeneratorConfig;
pub use generator::GeneratorError;
pub use generator::GeneratorUtils;
pub use generator::ModelInputsConfig;

Modules§

filter
Filters for processing model outputs prior to sampling.
generator
Tools to run the generation loop for an auto-regressive model.
metrics
Record timing metrics during generation.
model
Abstraction over rten::Model for querying and executing ML models.
sampler
Samplers which select a token from model outputs.
text_decoder
Iterator adapters to decode token IDs into text using rten-text.