tracel-ai/burn · error

Can load model checkpoint.

Error message

Can load model checkpoint.

What it means

Guard in `RLCheckpointer::load_checkpoint`: restoring the policy record for the requested epoch failed — typically the checkpoint file is missing or corrupt — so loading panics instead of returning a partially restored policy.

Source

Thrown at crates/burn-train/src/learner/rl/checkpointer.rs:65

                        .expect("Can save policy checkpoint.");
                    self.learning_agent
                        .save(epoch, learning_agent.record())
                        .expect("Can save learning agent checkpoint.");
                }
            }
        }
    }

    /// Load a training checkpoint.
    pub fn load_checkpoint(
        &self,
        learning_agent: RLC::LearningAgent,
        epoch: usize,
    ) -> RLC::LearningAgent {
        let record = self
            .policy
            .restore(epoch)
            .expect("Can load model checkpoint.");
        let policy = learning_agent.policy().load_record(record);

        let record = self
            .learning_agent
            .restore(epoch)
            .expect("Can load learning agent checkpoint.");
        let mut learning_agent = learning_agent.load_record(record);
        learning_agent.update_policy(policy);

        learning_agent
    }
}

View on GitHub (pinned to d16f7ba2ed)

Solutions

1. Verify a checkpoint for that epoch exists before calling load_checkpoint
2. Check the checkpoint file is not corrupt (matching burn version/features)
3. List available epochs via the file metadata to pick a valid one

When it happens

Trigger: Thrown at crates/burn-train/src/learner/rl/checkpointer.rs:65 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/be445cc23b4c1693. Report an issue: GitHub.