fix log epoch (#1986)

* fix logs * fix tests * remove dead loggers * fix tests * add prints * fix logs
mosaicml · Feb 22, 2023 · ddf179c · ddf179c
1 parent b16fb66
commit ddf179c
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 7 deletions.
diff --git a/composer/trainer/trainer.py b/composer/trainer/trainer.py
@@ -1921,7 +1921,7 @@ def _train_loop(self) -> None:
             try:
                 if int(self.state.timestamp.batch_in_epoch) == 0:
                     self.engine.run_event(Event.EPOCH_START)
-                    self.logger.log_metrics({'epoch': int(self.state.timestamp.epoch)})
+                    self.logger.log_metrics({'trainer/epoch': int(self.state.timestamp.epoch)})
 
                 dataloader = self.state.dataloader
                 if isinstance(dataloader, DataLoader) and isinstance(dataloader.sampler, DistributedSampler):
@@ -2760,9 +2760,6 @@ def _eval_loop(
 
                 self.engine.run_event(Event.EVAL_BATCH_END)
 
-            self.logger.log_metrics({'epoch': self.state.timestamp.epoch.value})
-            self.logger.log_metrics({'trainer/global_step': self.state.timestamp.batch.value})
-
             self._compute_and_log_metrics(dataloader_label=dataloader_label, metrics=metrics)
 
             self.engine.run_event(Event.EVAL_END)

diff --git a/tests/loggers/test_cometml_logger.py b/tests/loggers/test_cometml_logger.py
@@ -194,7 +194,8 @@ def test_comet_ml_logging_train_loop(monkeypatch, tmp_path):
 
     # Check that basic metrics appear in the comet logs
     assert len([
-        metric_msg for metric_msg in msg_type_to_msgs['metric_msg'] if metric_msg['metric']['metricName'] == 'epoch'
+        metric_msg for metric_msg in msg_type_to_msgs['metric_msg']
+        if metric_msg['metric']['metricName'] == 'trainer/epoch'
     ]) == 2
 
     # Check that basic params appear in the comet logs

diff --git a/tests/metrics/test_current_metrics.py b/tests/metrics/test_current_metrics.py
@@ -110,13 +110,12 @@ def test_current_metrics(eval_interval: str,):
     num_expected_calls += (train_subset_num_batches + 1) * num_epochs
     # computed at eval end
     if compute_val_metrics:
-        num_calls_per_eval = 3  # metrics + epoch + trainer/global_step
         num_evals = 0
         if eval_interval == '1ba':
             num_evals += train_subset_num_batches * num_epochs
         if eval_interval == '1ep':
             num_evals += num_epochs
-        num_expected_calls += (num_calls_per_eval) * num_evals
+        num_expected_calls += num_evals
 
     num_actual_calls = len(mock_logger_destination.log_metrics.mock_calls)