tracel-ai/burn · error

Default content should be implemented for {self_type}.

Error message

Default content should be implemented for {self_type}.

What it means

Module display formatting first tries `custom_content`, and when that returns None it falls back to `content()` (the default content renderer). This panic fires when a module (or its primitive) implements the display traits but returns None from `content`, i.e. the required default renderer is missing. It is a trait-implementation contract violation, not a runtime data problem.

Source

Thrown at crates/burn-core/src/module/display.rs:65

        } else {
            passed_settings
        };

        let indent = " ".repeat(settings.level * settings.indentation_size());
        let indent_close_braces = " ".repeat((settings.level - 1) * settings.indentation_size());

        let settings = settings.level_up();

        let self_type = extract_type_name::<Self>();

        // Use custom content if it is implemented and show_all_attributes is false,
        // otherwise use default content
        let content = if !settings.show_all_attributes() {
            self.custom_content(Content::new(settings.clone()))
                .unwrap_or_else(|| {
                    self.content(Content::new(settings.clone()))
                        .unwrap_or_else(|| {
                            panic!("Default content should be implemented for {self_type}.")
                        })
                })
        } else {
            self.content(Content::new(settings.clone()))
                .unwrap_or_else(|| panic!("Default content should be implemented for {self_type}."))
        };

        let top_level_type = if let Some(top_level_type) = content.top_level_type {
            top_level_type.to_owned()
        } else {
            self_type.to_owned()
        };

        // If there is only one item in the content, return it or no attributes
        if let Some(item) = content.single_item {
            return item;
        } else if content.attributes.is_empty() {
            return top_level_type.to_string();

View on GitHub (pinned to d16f7ba2ed)

Solutions

1. Implement `content(...)` for the type named in {self_type} so it returns Some(Content) with at least the type name and parameters.
2. If the type intentionally has no content, wrap rendering so `content` returns a minimal Content instead of None.
3. Check the burn version's display trait requirements — an upgrade may have made `content` mandatory for your custom module.

Example fix

// before
impl ModuleContent for MyLayer {
    fn content(&self, _settings: &ContentSettings) -> Option<Content> { None } // panics
}
// after
impl ModuleContent for MyLayer {
    fn content(&self, settings: &ContentSettings) -> Option<Content> {
        Some(Content::new(settings.clone())
            .add_top_level_type("MyLayer")
            .add_parameter(Param::new("in_features", self.in_features)))
    }
}

// Before printing, assert your custom modules implement content
fn check_content<M: ModuleContent>(m: &M, s: &ContentSettings) {
    assert!(m.content(Content::new(s.clone())).is_some(), "content() must be implemented");
}

When it happens

Trigger: Implementing `ModuleDisplay`/`ModuleContent` (or a new primitive/backend tensor display) whose `content()` returns `None` while `show_all_attributes()` is false; a custom module that only overrides `custom_content` and never supplies default content; `format` called on a type whose content impl was forgotten after an API update.

Common situations: Printing/logging a custom module with `model.print()` or `{:#?}`-style display after upgrading burn and missing a new required `content` implementation; hand-written adapters around burn modules returning None unconditionally.

Related errors

• Can't format yet
• capture tensor operations must run inside CaptureDevice::cap
• capture tensor {} has no initialized value
• seeding is not supported during graph capture
• capture graph {graph_id:?} was not registered

AI-assisted analysis of tracel-ai/burn@d16f7ba2ed (2026-09-05). Data as JSON: /api/errors/acba8e248713ba35. Report an issue: GitHub.