pub struct ContextAgent<A: Agent> { /* private fields */ }Expand description
Context agent wraps another agent and provides context-aware responses
Implementations§
Source§impl<A: Agent> ContextAgent<A>
impl<A: Agent> ContextAgent<A>
Sourcepub fn new(base_agent: A, max_context_items: usize) -> Self
pub fn new(base_agent: A, max_context_items: usize) -> Self
Creates a new context agent with a base agent and maximum context items
Sourcepub fn with_context_format(self, format: ContextFormat) -> Self
pub fn with_context_format(self, format: ContextFormat) -> Self
Sets the context format for generating context strings
Trait Implementations§
Source§impl<A: Agent> Agent for ContextAgent<A>
impl<A: Agent> Agent for ContextAgent<A>
Source§fn predict(&self, input: &str) -> ResponseFormat
fn predict(&self, input: &str) -> ResponseFormat
Generates a response with context added to the input
Source§fn predict_with_metadata(&self, input: &str) -> PredictionResult
fn predict_with_metadata(&self, input: &str) -> PredictionResult
Erweiterte Vorhersage mit Metadaten und Konfidenz
Source§fn confidence(&self, input: &str) -> f32
fn confidence(&self, input: &str) -> f32
Bestimmt die Konfidenz des Agenten für eine bestimmte Eingabe (0.0 - 1.0)
Source§fn can_answer(&self, input: &str) -> bool
fn can_answer(&self, input: &str) -> bool
Prüft, ob der Agent die Eingabe beantworten kann
Source§fn predict_text(&self, input: &str) -> String
fn predict_text(&self, input: &str) -> String
Hilfsmethode für Rückwärtskompatibilität
Source§impl<A: Agent> ContextualAgent for ContextAgent<A>
impl<A: Agent> ContextualAgent for ContextAgent<A>
Source§fn add_context(&mut self, question: String, answer: ResponseFormat)
fn add_context(&mut self, question: String, answer: ResponseFormat)
Adds a new context item to the conversation history
Source§fn clear_context(&mut self)
fn clear_context(&mut self)
Clears the entire context history
Source§fn add_text_context(&mut self, question: String, answer: String)
fn add_text_context(&mut self, question: String, answer: String)
Hilfsmethode für Textantworten
Source§impl<A: TrainableAgent> TrainableAgent for ContextAgent<A>
impl<A: TrainableAgent> TrainableAgent for ContextAgent<A>
Source§fn train(&mut self, data: &[TrainingExample])
fn train(&mut self, data: &[TrainingExample])
Trains the base agent with the provided training data (replaces all)
Source§fn append(&mut self, data: &[TrainingExample])
fn append(&mut self, data: &[TrainingExample])
Appends training data to the base agent without replacing
Source§fn train_single(&mut self, example: &TrainingExample)
fn train_single(&mut self, example: &TrainingExample)
Fügt ein einzelnes Beispiel hinzu (ohne vorherige Daten zu ersetzen)
Source§fn train_legacy(&mut self, data: &[LegacyTrainingExample])
fn train_legacy(&mut self, data: &[LegacyTrainingExample])
Hilfsmethode für das Training mit Legacy-Daten
Source§fn add_example(
&mut self,
input: &str,
output: impl Into<ResponseFormat>,
weight: f32,
)
fn add_example( &mut self, input: &str, output: impl Into<ResponseFormat>, weight: f32, )
Fügt ein neues Trainingsbeispiel hinzu (ohne vorherige Daten zu ersetzen)
Auto Trait Implementations§
impl<A> Freeze for ContextAgent<A>where
A: Freeze,
impl<A> !RefUnwindSafe for ContextAgent<A>
impl<A> Send for ContextAgent<A>where
A: Send,
impl<A> Sync for ContextAgent<A>where
A: Sync,
impl<A> Unpin for ContextAgent<A>where
A: Unpin,
impl<A> UnsafeUnpin for ContextAgent<A>where
A: UnsafeUnpin,
impl<A> !UnwindSafe for ContextAgent<A>
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more