Struct RftHyperParameters 

#[non_exhaustive]
pub struct RftHyperParameters {
    pub epoch_count: Option<i32>,
    pub batch_size: Option<i32>,
    pub learning_rate: Option<f32>,
    pub max_prompt_length: Option<i32>,
    pub training_sample_per_prompt: Option<i32>,
    pub inference_max_tokens: Option<i32>,
    pub reasoning_effort: Option<ReasoningEffort>,
    pub eval_interval: Option<i32>,
}

Hyperparameters for controlling the reinforcement fine-tuning training process, including learning settings and evaluation intervals.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

epoch_count: Option<i32>

Number of training epochs to run during reinforcement fine-tuning. Higher values may improve performance but increase training time.

batch_size: Option<i32>

Number of training samples processed in each batch during reinforcement fine-tuning (RFT) training. Larger batches may improve training stability.

learning_rate: Option<f32>

Learning rate for the reinforcement fine-tuning. Controls how quickly the model adapts to reward signals.

max_prompt_length: Option<i32>

Maximum length of input prompts during RFT training, measured in tokens. Longer prompts allow more context but increase memory usage and training-time.

training_sample_per_prompt: Option<i32>

Number of response samples generated per prompt during RFT training. More samples provide better reward signal estimation.

inference_max_tokens: Option<i32>

Maximum number of tokens the model can generate in response to each prompt during RFT training.

reasoning_effort: Option<ReasoningEffort>

Level of reasoning effort applied during RFT training. Higher values may improve response quality but increase training time.

eval_interval: Option<i32>

Interval between evaluation runs during RFT training, measured in training steps. More frequent evaluation provides better monitoring.

Implementations

impl RftHyperParameters

pub fn epoch_count(&self) -> Option<i32>

Number of training epochs to run during reinforcement fine-tuning. Higher values may improve performance but increase training time.

pub fn batch_size(&self) -> Option<i32>

Number of training samples processed in each batch during reinforcement fine-tuning (RFT) training. Larger batches may improve training stability.

pub fn learning_rate(&self) -> Option<f32>

Learning rate for the reinforcement fine-tuning. Controls how quickly the model adapts to reward signals.

pub fn max_prompt_length(&self) -> Option<i32>

Maximum length of input prompts during RFT training, measured in tokens. Longer prompts allow more context but increase memory usage and training-time.

pub fn training_sample_per_prompt(&self) -> Option<i32>

Number of response samples generated per prompt during RFT training. More samples provide better reward signal estimation.

pub fn inference_max_tokens(&self) -> Option<i32>

Maximum number of tokens the model can generate in response to each prompt during RFT training.

pub fn reasoning_effort(&self) -> Option<&ReasoningEffort>

Level of reasoning effort applied during RFT training. Higher values may improve response quality but increase training time.

pub fn eval_interval(&self) -> Option<i32>

Interval between evaluation runs during RFT training, measured in training steps. More frequent evaluation provides better monitoring.

impl RftHyperParameters

pub fn builder() -> RftHyperParametersBuilder

Creates a new builder-style object to manufacture RftHyperParameters.

Trait Implementations

impl Clone for RftHyperParameters

fn clone(&self) -> RftHyperParameters

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for RftHyperParameters

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl PartialEq for RftHyperParameters

fn eq(&self, other: &RftHyperParameters) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for RftHyperParameters