New train profile argument for loggers (#2862)

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

ultralytics/yolo/utils/callbacks/__init__.py

@@ -1,3 +1,5 @@
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+
 from .base import add_integration_callbacks, default_callbacks, get_default_callbacks
 
 __all__ = 'add_integration_callbacks', 'default_callbacks', 'get_default_callbacks'

ultralytics/yolo/utils/callbacks/clearml.py

@@ -1,11 +1,12 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
+
 import re
 
 import matplotlib.image as mpimg
 import matplotlib.pyplot as plt
 
 from ultralytics.yolo.utils import LOGGER, TESTS_RUNNING
-from ultralytics.yolo.utils.torch_utils import get_flops, get_num_params
+from ultralytics.yolo.utils.torch_utils import model_info_for_loggers
 
 try:
     import clearml
@@ -105,11 +106,7 @@ def on_fit_epoch_end(trainer):
                                         value=trainer.epoch_time,
                                         iteration=trainer.epoch)
         if trainer.epoch == 0:
-            model_info = {
-                'model/parameters': get_num_params(trainer.model),
-                'model/GFLOPs': round(get_flops(trainer.model), 3),
-                'model/speed(ms)': round(trainer.validator.speed['inference'], 3)}
-            for k, v in model_info.items():
+            for k, v in model_info_for_loggers(trainer).items():
                 task.get_logger().report_single_value(k, v)
 
 

ultralytics/yolo/utils/callbacks/comet.py

@@ -1,9 +1,10 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
+
 import os
 from pathlib import Path
 
 from ultralytics.yolo.utils import LOGGER, RANK, TESTS_RUNNING, ops
-from ultralytics.yolo.utils.torch_utils import get_flops, get_num_params
+from ultralytics.yolo.utils.torch_utils import model_info_for_loggers
 
 try:
     import comet_ml
@@ -324,11 +325,7 @@ def on_fit_epoch_end(trainer):
     experiment.log_metrics(trainer.metrics, step=curr_step, epoch=curr_epoch)
     experiment.log_metrics(trainer.lr, step=curr_step, epoch=curr_epoch)
     if curr_epoch == 1:
-        model_info = {
-            'model/parameters': get_num_params(trainer.model),
-            'model/GFLOPs': round(get_flops(trainer.model), 3),
-            'model/speed(ms)': round(trainer.validator.speed['inference'], 3), }
-        experiment.log_metrics(model_info, step=curr_step, epoch=curr_epoch)
+        experiment.log_metrics(model_info_for_loggers(trainer), step=curr_step, epoch=curr_epoch)
 
     if not save_assets:
         return

ultralytics/yolo/utils/callbacks/hub.py

@@ -5,7 +5,7 @@ from time import time
 
 from ultralytics.hub.utils import PREFIX, events
 from ultralytics.yolo.utils import LOGGER
-from ultralytics.yolo.utils.torch_utils import get_flops, get_num_params
+from ultralytics.yolo.utils.torch_utils import model_info_for_loggers
 
 
 def on_pretrain_routine_end(trainer):
@@ -24,11 +24,7 @@ def on_fit_epoch_end(trainer):
         # Upload metrics after val end
         all_plots = {**trainer.label_loss_items(trainer.tloss, prefix='train'), **trainer.metrics}
         if trainer.epoch == 0:
-            model_info = {
-                'model/parameters': get_num_params(trainer.model),
-                'model/GFLOPs': round(get_flops(trainer.model), 3),
-                'model/speed(ms)': round(trainer.validator.speed['inference'], 3)}
-            all_plots = {**all_plots, **model_info}
+            all_plots = {**all_plots, **model_info_for_loggers(trainer)}
         session.metrics_queue[trainer.epoch] = json.dumps(all_plots)
         if time() - session.timers['metrics'] > session.rate_limits['metrics']:
             session.upload_metrics()

ultralytics/yolo/utils/callbacks/neptune.py

@@ -1,9 +1,10 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
+
 import matplotlib.image as mpimg
 import matplotlib.pyplot as plt
 
 from ultralytics.yolo.utils import LOGGER, TESTS_RUNNING
-from ultralytics.yolo.utils.torch_utils import get_flops, get_num_params
+from ultralytics.yolo.utils.torch_utils import model_info_for_loggers
 
 try:
     import neptune
@@ -68,11 +69,7 @@ def on_train_epoch_end(trainer):
 def on_fit_epoch_end(trainer):
     """Callback function called at end of each fit (train+val) epoch."""
     if run and trainer.epoch == 0:
-        model_info = {
-            'parameters': get_num_params(trainer.model),
-            'GFLOPs': round(get_flops(trainer.model), 3),
-            'speed(ms)': round(trainer.validator.speed['inference'], 3)}
-        run['Configuration/Model'] = model_info
+        run['Configuration/Model'] = model_info_for_loggers(trainer)
     _log_scalars(trainer.metrics, trainer.epoch + 1)
 
 

ultralytics/yolo/utils/callbacks/raytune.py

@@ -1,3 +1,5 @@
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+
 try:
     import ray
     from ray import tune

ultralytics/yolo/utils/callbacks/tensorboard.py

@@ -1,4 +1,5 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
+
 from ultralytics.yolo.utils import LOGGER, TESTS_RUNNING, colorstr
 
 try:

ultralytics/yolo/utils/callbacks/wb.py

@@ -1,30 +1,27 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-from ultralytics.yolo.utils.torch_utils import get_flops, get_num_params
+from ultralytics.yolo.utils import TESTS_RUNNING
+from ultralytics.yolo.utils.torch_utils import model_info_for_loggers
 
 try:
     import wandb as wb
 
     assert hasattr(wb, '__version__')
+    assert not TESTS_RUNNING  # do not log pytest
 except (ImportError, AssertionError):
     wb = None
 
 
 def on_pretrain_routine_start(trainer):
     """Initiate and start project if module is present."""
-    wb.init(project=trainer.args.project or 'YOLOv8', name=trainer.args.name, config=vars(
-        trainer.args)) if not wb.run else wb.run
+    wb.run or wb.init(project=trainer.args.project or 'YOLOv8', name=trainer.args.name, config=vars(trainer.args))
 
 
 def on_fit_epoch_end(trainer):
     """Logs training metrics and model information at the end of an epoch."""
     wb.run.log(trainer.metrics, step=trainer.epoch + 1)
     if trainer.epoch == 0:
-        model_info = {
-            'model/parameters': get_num_params(trainer.model),
-            'model/GFLOPs': round(get_flops(trainer.model), 3),
-            'model/speed(ms)': round(trainer.validator.speed['inference'], 3)}
-        wb.run.log(model_info, step=trainer.epoch + 1)
+        wb.run.log(model_info_for_loggers(trainer), step=trainer.epoch + 1)
 
 
 def on_train_epoch_end(trainer):