llm_chain::tokens

Trait ExecutorTokenCountExt

Source
pub trait ExecutorTokenCountExt: Executor {
    // Provided method
    fn split_to_fit(
        &self,
        step: &Step,
        doc: &Parameters,
        base_parameters: &Parameters,
        chunk_overlap: Option<usize>,
    ) -> Result<Vec<Parameters>, PromptTokensError> { ... }
}
Expand description

An extension trait for the Executor trait that provides additional methods for working with token counts.

Provided Methods§

Source

fn split_to_fit( &self, step: &Step, doc: &Parameters, base_parameters: &Parameters, chunk_overlap: Option<usize>, ) -> Result<Vec<Parameters>, PromptTokensError>

Splits a Parameters object into multiple smaller Parameters objects that fit within the context window size supported by the given model.

§Arguments
  • step - The step that will process the Parameters. Has impact on tokenizer & text splitter used
  • doc - The parameter object to split into multiple, smaller, parameters
  • chunk_overlap - The amount of tokens each split part should overlap with previous & next chunk
§Errors

Returns a PromptTokensError if there is an issue computing the tokens.

Dyn Compatibility§

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.

Implementors§

Source§

impl<E: Executor> ExecutorTokenCountExt for E

Blanket implementation of ExecutorTokenCountExt for all Executors