Rich 14.1.0+ causes IndexError in training progress bar

[ashinkajay]

Bug description

With version 14.1.0 of rich library the progress bar loaded during training is throwing error. The error doesn't occur when rich library is of version 14.0.0.

Error trace added below.

What version are you seeing the problem on?

v2.2

Reproduced in studio

No response

How to reproduce the bug

Error messages and logs

                                                                                                           
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:543: in fit                                                                                 
    call._call_and_handle_interrupt(                                                                                                                                                          
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py:44: in _call_and_handle_interrupt                                                              
    return trainer_fn(*args, **kwargs)                                                                                                                                                        
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:579: in _fit_impl                                                                           
    self._run(model, ckpt_path=ckpt_path)                                                                                                                                                     
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:986: in _run                                                                                
    results = self._run_stage()                                                                                                                                                               
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:1032: in _run_stage                                                                         
    self.fit_loop.run()                                                                                                                                                                       
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/loops/fit_loop.py:201: in run                                                                                  
    self.on_run_start()                                                                                                                                                                       
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/loops/fit_loop.py:327: in on_run_start                                                                         
    call._call_callback_hooks(trainer, "on_train_start")                                                                                                                                      
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py:208: in _call_callback_hooks                                                                   
    fn(trainer, trainer.lightning_module, *args, **kwargs)                                                                                                                                    
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/callbacks/progress/rich_progress.py:368: in on_train_start                                                     
    self._init_progress(trainer)                                                                                                                                                              
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/pytorch_lightning/callbacks/progress/rich_progress.py:344: in _init_progress                                                     
    self._console.clear_live()                                                                                                                                                                
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
                                                                                                                                                                                              
self = <console width=190 None>                                                                                                                                                               
                                                                                                                                                                                              
    def clear_live(self) -> None:                                                                                                                                                             
        """Clear the Live instance. Used by the Live context manager (no need to call directly)."""                                                                                           
        with self._lock:                                                                                                                                                                      
>           self._live_stack.pop()                                                                                                                                                            
E           IndexError: pop from empty list                                                                                                                                                   
                                                                                                                                                                                              
/home/user/miniconda3/envs/tts/lib/python3.10/site-packages/rich/console.py:847: IndexError   

Environment

Current environment

- PyTorch Lightning Version (e.g., 2.5.0): 2.2.0
- PyTorch Version (e.g., 2.5): 2.2.1+cu121
- Python version (e.g., 3.12): 3.10.16
- OS (e.g., Linux): Linux
- How you installed Lightning(`conda`, `pip`, source): pip

More info

No response

[deependujha]

Hi @ashinkajay,

Thanks for reporting this!

This issue is already resolved in the master branch via #21016. It was caused by a breaking change in rich >=14.1.0, where self._live_stack.pop() could raise an IndexError if the stack is empty.

Workaround:

Until a new release is published, you can either:

• Use the master branch of PyTorch Lightning,

pip install git+https://github.com/Lightning-AI/pytorch-lightning.git

• Or, Downgrade rich to 14.0.0 as a temporary workaround:

  pip install rich==14.0.0

Let us know if you encounter any further issues!

[SkafteNicki]

PR #21016 was released in v2.5.3 https://github.com/Lightning-AI/pytorch-lightning/releases/tag/2.5.3 yesterday.
Please update to the new version. Closing issue.