maxent\_grpo.training.metrics
=============================

.. automodule:: maxent_grpo.training.metrics

   
   .. rubric:: Functions

   .. autosummary::
   
      _as_float
      _base_metric_block
      _build_metrics_payload
      _build_sample_table
      _clip_metric_block
      _drop_prefix
      _emit_metrics
      _entropy_bonus_impact
      _epoch_from_global_step
      _filter_metrics
      _fraction_zero_std_groups
      _gather_dict_of_lists_for_metrics
      _gather_list_for_metrics
      _get_wandb
      _length_metric_block
      _log_debug_metrics
      _log_entropy_bonus_impact
      _log_like_grpo_enabled
      _log_sample_table
      _logging_controls
      _loss_component_block
      _mean_std
      _metrics_mode
      _policy_entropy_from_scores
      _pretty_print_metrics
      _quantile_stats
      _reward_component_stats
      _reward_metric_block
      _rich_completion_sync_enabled
      _rich_completion_wandb_enabled
      _should_log
      _slim_metrics
      _sum_scalar_for_metrics
      _summarize_reward_stats
      _summarize_weight_stats
      _update_weighting_history
      _wait_after_rich_completion_logging
      _weight_metric_block
      _weighting_config_block
      _write_sample_table_sidecar
      accumulate_metrics
      build_training_metrics_dict
      flush_metric_averages
      log_local_step
      log_training_metrics
      log_training_step
      summarize_reward_stats
      summarize_weight_stats
   
   .. rubric:: Exceptions

   .. autosummary::
   
      _FallbackWandbError