tracel-ai/burn · error

Killing training from user input.

Error message

Killing training from user input.

What it means

Deliberate shutdown panic in the TUI renderer's `send_event`: a kill was requested (via the TUI kill signal channel or the global USER_KILL_REQUESTED flag), so the renderer panics to tear down the rendering thread and stop training from user input.

Source

Thrown at crates/burn-train/src/renderer/tui/renderer.rs:135

        let init = Progress::new(0, 0, None);
        Self {
            sender,
            interrupter,
            handle_join: Some(handle_join),
            kill_signal: Arc::new(Mutex::new(kill_signal_receiver)),
            current_split: TuiSplit::Train,
            training_progress: ProgressSnapshot::new(init.clone(), init.clone()),
            eval_progress: ProgressSnapshot::new(init.clone(), init),
        }
    }

    fn send_event(&self, event: TuiRendererEvent) {
        if self.kill_signal.lock().unwrap().try_recv().is_ok()
            || USER_KILL_REQUESTED.load(Ordering::Relaxed)
        {
            USER_KILL_REQUESTED.store(true, Ordering::Relaxed);
            panic!("Killing training from user input.");
        }

        if let Err(e) = self.sender.send(event) {
            // Ignore send errors if we are already in the middle of killing
            if !USER_KILL_REQUESTED.load(Ordering::Relaxed) {
                log::warn!("Failed to send TUI event: {e}");
            }
        }
    }

    /// Set the renderer to persistent mode.
    pub fn persistent(self) -> Self {
        self.send_event(TuiRendererEvent::Persistent);
        self
    }
}

struct TuiMetricsRenderer {

View on GitHub (pinned to d16f7ba2ed)

Solutions

1. This is an intentional user-initiated termination; no fix is needed
2. Catch the unwind in the training harness if you need cleanup before exit
3. Use the interrupter/graceful-stop API instead of the kill signal when a soft stop is desired

When it happens

Trigger: Thrown at crates/burn-train/src/renderer/tui/renderer.rs:135 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tracel-ai/burn@d16f7ba2ed (2026-09-05). Data as JSON: /api/errors/27f309ba2a51804c. Report an issue: GitHub.