Skip to content

PyTorch Lightning + torch.compile: MisconfigurationException on self.log

Training a Lightning model with torch.compile crashes in the logger connector with a 'called self.log twice with different arguments' MisconfigurationException raised inside the compiled region. Dynamo traces through Lightning's result collection and the metadata-equality branch raises instead of breaking the graph. Insert a graph break or compile only the model forward.

Quick answer

Training a Lightning model with torch.

Training Stability#pytorch-lightning#torch-compile#dynamo#graph-break#logging#training-stability

What this failure is

PyTorch Lightning + torch.compile: MisconfigurationException on self.log is a Training Stability failure seen during ML training runs. Training a Lightning model with torch.compile crashes in the logger connector with a 'called self.log twice with different arguments' MisconfigurationException raised inside the compiled region. Dynamo traces through Lightning's result collection and the metadata-equality branch raises instead of breaking the graph. Insert a graph break or compile only the model forward. Common tags: Pytorch Lightning, Torch Compile, Dynamo, Graph Break.

Live diagnosis, no signup

Is this what broke your run? Paste your log.

You're reading about PyTorch Lightning + torch.compile: MisconfigurationException on self.log. Paste your own crash log or traceback below and get the real root cause for YOUR run, not this generic entry. No account, no card. Logs are masked at ingress and never saved to account history.

training_logs.txt
No log to hand? Try one:

3 free diagnoses/day

Want 14 days on the Scale plan?

Request an evaluation code. A verified workplace organization activates up to 50 diagnoses a day, alerts, history, and follow-up questions. No credit card or automatic subscription.

Evaluate one incident

Why it happens (the mechanism)

Torch.compile/Dynamo traces through Lightning's _ResultCollection logging path. The metadata-equality check raises inside the compiled graph instead of being a normal control-flow branch, so a harmless re-log turns into an exception. Taken together, these mechanisms explain why the failure is reproducible, why it tends to surface on specific workloads or scales, and why generic mitigation attempts often fall short without addressing the underlying cause.

What you'll observe

  • torch.compile + Lightning crashes during logging
  • A benign re-log path becomes a hard exception
  • Removing torch.compile makes the error disappear

Common symptoms and what they mean

SymptomWhy it happens
MisconfigurationException: You called self.log(...) twice in ... with different argumentstorch.compile/Dynamo traces through Lightning's _ResultCollection logging path
Traceback through logger_connector/result.py _ResultCollectionThe metadata-equality check raises inside the compiled graph instead of being a normal control-flow branch, so a harmless re-log turns into an exception
Only occurs with torch.compile enabledtorch.compile/Dynamo traces through Lightning's _ResultCollection logging path

Which systems are affected

  • PyTorch Lightning 2.x + torch.compile
  • Compiling the LightningModule/training loop (not just the inner model)
  • self.log calls inside compiled forward/step

How to confirm this is the problem

Use this checklist to test the hypothesis against a small reproduction. No single line proves the root cause, so preserve the preceding events and compare one variable at a time.

  • Reproduce the failure from a clean checkpoint/seed: the symptom must appear without warm-up state from a previous run.
  • Verified signal present: MisconfigurationException: You called self.log(...) twice in ... with different arguments
  • Verified signal present: Traceback through logger_connector/result.py _ResultCollection
  • Verified signal present: Only occurs with torch.compile enabled
  • A targeted fix from the "How to fix it" section eliminates or substantially reduces the symptom within one validation pass.

The fix and the prevention pattern

The root cause is on this page and stays free. A free account adds the exact remediation steps, saved history, and the fix on every entry in the encyclopedia.

Sign up free. Unlock the full analysis

No credit card. Daily allowance follows verified trust tier. Instant access.

Diagnose this failure in VS Code

Select the traceback or open the failed terminal, then run Denpex locally to see the initiating rank, collateral failures, exact fix, and verification command without uploading the log.

Install the free VS Code extension

Root cause

  • torch.compile/Dynamo traces through Lightning's _ResultCollection logging path
  • The metadata-equality check raises inside the compiled graph instead of being a normal control-flow branch, so a harmless re-log turns into an exception

The fix and how to prevent it

Evaluate Denpex on your own logs

Request a Scale evaluation code. A verified workplace organization activates 14 days with up to 50 diagnoses a day. Every account keeps its current diagnosis allowance and gets a verification path. No card or automatic subscription.

We send a single-use code tied to that address. Static provider and TLD rules do not reject valid addresses. Account trust determines the benefit after signup.

Don't just read the fix, diagnose your run

The encyclopedia tells you what went wrong. Denpex tells you what went wrong in YOUR training run. With your logs, your config, and your stack.