Struct TestGenCorpus 

pub struct TestGenCorpus {
    pub train: Vec<TestGenSample>,
    pub validation: Vec<TestGenSample>,
    pub test: Vec<TestGenSample>,
}

Test generation corpus

Fields

train: Vec<TestGenSample>

Training samples

validation: Vec<TestGenSample>

Validation samples

test: Vec<TestGenSample>

Test samples (holdout)

Implementations

impl TestGenCorpus

pub const fn new() -> Self

Create empty corpus

pub fn load_jsonl( train_path: &Path, validation_path: &Path, test_path: &Path, ) -> Result<Self, CorpusError>

Load corpus from JSONL files

Errors

Returns error if files cannot be read or parsed.

pub fn mock(train_size: usize, val_size: usize, test_size: usize) -> Self

Create mock corpus for testing

pub fn stats(&self) -> CorpusStats

Get corpus statistics

pub fn len(&self) -> usize

Total number of samples

pub fn is_empty(&self) -> bool

Check if corpus is empty

pub fn shuffle_train(&mut self, seed: u64)

Shuffle training data with seed

pub fn format_prompt(sample: &TestGenSample) -> String

Format sample as prompt for model

pub fn format_target(sample: &TestGenSample) -> String

Format sample as target for training

Trait Implementations

impl Clone for TestGenCorpus

fn clone(&self) -> TestGenCorpus

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TestGenCorpus

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

Formats the value using the given formatter.

impl Default for TestGenCorpus

fn default() -> Self

Returns the “default value” for a type.