{"record":{"id":"06648525bbe7f719","repo":"tracel-ai/burn","slug":"can-save-policy-checkpoint","errorCode":null,"errorMessage":"Can save policy checkpoint.","messagePattern":"Can save policy checkpoint\\.","errorType":"panic","errorClass":null,"httpStatus":null,"severity":"error","filePath":"crates/burn-train/src/learner/rl/checkpointer.rs","lineNumber":47,"sourceCode":"        epoch: usize,\n        store: &EventStoreClient,\n    ) {\n        let actions = self.strategy.checkpointing(epoch, store);\n\n        for action in actions {\n            match action {\n                CheckpointingAction::Delete(epoch) => {\n                    self.policy\n                        .delete(epoch)\n                        .expect(\"Can delete policy checkpoint.\");\n                    self.learning_agent\n                        .delete(epoch)\n                        .expect(\"Can delete learning agent checkpoint.\")\n                }\n                CheckpointingAction::Save => {\n                    self.policy\n                        .save(epoch, policy.clone().into_record())\n                        .expect(\"Can save policy checkpoint.\");\n                    self.learning_agent\n                        .save(epoch, learning_agent.record())\n                        .expect(\"Can save learning agent checkpoint.\");\n                }\n            }\n        }\n    }\n\n    /// Load a training checkpoint.\n    pub fn load_checkpoint(\n        &self,\n        learning_agent: RLC::LearningAgent,\n        epoch: usize,\n    ) -> RLC::LearningAgent {\n        let record = self\n            .policy\n            .restore(epoch)\n            .expect(\"Can load model checkpoint.\");","sourceCodeStart":29,"sourceCodeEnd":65,"githubUrl":"https://github.com/tracel-ai/burn/blob/d16f7ba2ed0d41408189384044cc886fb4c8f957/crates/burn-train/src/learner/rl/checkpointer.rs#L29-L65","documentation":"Guard in the RL checkpointer's `checkpoint`: saving the policy checkpoint for the current epoch failed (I/O, serialization, or missing directory), so the checkpointer panics rather than continue training without a durable policy snapshot.","triggerScenarios":"Thrown at crates/burn-train/src/learner/rl/checkpointer.rs:47 when the library encounters an invalid state.","commonSituations":"See trigger scenarios.","solutions":["Check disk space and write permissions on the checkpoint directory","Inspect the inner save error (serialization/path issues)","Create the checkpoint directory before training starts"],"exampleFix":null,"handlingStrategy":"retry","validationCode":null,"typeGuard":null,"tryCatchPattern":null,"preventionTips":[],"tags":[],"backgroundTag":null,"analyzedSha":"d16f7ba2ed0d41408189384044cc886fb4c8f957","analyzedAt":"2026-09-05T13:19:14.260Z","contentChangedAt":"2026-09-05T13:19:14.260Z","schemaVersion":2},"datasetVersion":"2026-09-12T17:17:11.597Z"}