Cherrypick: Cleanup: move to per-module loggers instead of the global…

… logging object (#2539)
ludwig-ai · Sep 20, 2022 · c2f4e42 · c2f4e42
1 parent 26d707b
commit c2f4e42
Show file tree

Hide file tree

Showing 25 changed files with 133 additions and 100 deletions.
diff --git a/README.md b/README.md
@@ -409,4 +409,4 @@ know, please consider [joining the Ludwig Slack](https://join.slack.com/t/ludwig
 - [Slack](https://join.slack.com/t/ludwig-ai/shared_invite/zt-mrxo87w6-DlX5~73T2B4v_g6jj0pJcQ)
 - [Twitter](https://twitter.com/ludwig_ai)
 - [Medium](https://medium.com/ludwig-ai)
-- [GitHub Issues](https://github.com/ludwig-ai/ludwig/issues)
+- [GitHub Issues](https://github.com/ludwig-ai/ludwig/issues)
diff --git a/ludwig/automl/auto_tune_config.py b/ludwig/automl/auto_tune_config.py
@@ -35,6 +35,8 @@
 from ludwig.utils.defaults import merge_with_defaults
 from ludwig.utils.torch_utils import initialize_pytorch
 
+logger = logging.getLogger(__name__)
+
 # maps variable search space that can be modified to minimum permissible value for the range
 RANKED_MODIFIABLE_PARAM_LIST = {
     "tabnet": OrderedDict(
@@ -228,7 +230,7 @@ def memory_tune_config(config, dataset, model_category, row_count, backend):
             if reduce_text_feature_max_length(config, training_set_metadata):
                 reduce_text_feature_max_length(temp_config, training_set_metadata)
                 mem_use = compute_memory_usage(temp_config, training_set_metadata, model_category)
-        logging.info(f"Checking model estimated mem use {mem_use} against memory size {max_memory}")
+        logger.info(f"Checking model estimated mem use {mem_use} against memory size {max_memory}")
         if mem_use <= max_memory:
             fits_in_memory = True
             break

diff --git a/ludwig/automl/automl.py b/ludwig/automl/automl.py
@@ -58,6 +58,8 @@
     raise ImportError(" ray is not installed. In order to use auto_train please run pip install ludwig[ray]")
 
 
+logger = logging.getLogger(__name__)
+
 OUTPUT_DIR = "."
 
 
@@ -77,7 +79,7 @@ def best_trial_id(self) -> str:
     def best_model(self) -> Optional[LudwigModel]:
         checkpoint = self._experiment_analysis.best_checkpoint
         if checkpoint is None:
-            logging.warning("No best model found")
+            logger.warning("No best model found")
             return None
 
         if not _ray_113:

diff --git a/ludwig/backend/datasource.py b/ludwig/backend/datasource.py
@@ -110,7 +110,7 @@ def _read_file(
             try:
                 data = get_bytes_obj_from_http_path(path)
             except urllib3.exceptions.HTTPError as e:
-                logging.warning(e)
+                logger.warning(e)
                 data = None
         else:
             super_result = super()._read_file(f, path, **reader_args)[0]

diff --git a/ludwig/benchmarking/benchmark.py b/ludwig/benchmarking/benchmark.py
@@ -11,6 +11,7 @@
 from ludwig.utils.data_utils import load_yaml
 
 os.environ["TOKENIZERS_PARALLELISM"] = "false"
+logger = logging.getLogger(__name__)
 
 
 def setup_experiment(experiment: Dict[str, str]) -> Dict[Any, Any]:
@@ -32,7 +33,7 @@ def benchmark_one_local(experiment: Dict[str, str], export_artifacts_dict: Dict[
     experiment: dictionary containing the dataset name, config path, and experiment name.
     export_artifacts_dict: dictionary containing an export boolean flag and a path to export to.
     """
-    logging.info(f"\nRunning experiment *{experiment['experiment_name']}* on dataset *{experiment['dataset_name']}*")
+    logger.info(f"\nRunning experiment *{experiment['experiment_name']}* on dataset *{experiment['dataset_name']}*")
 
     # configuring backend and paths
     model_config = setup_experiment(experiment)

diff --git a/ludwig/benchmarking/summary_dataclasses.py b/ludwig/benchmarking/summary_dataclasses.py
@@ -11,6 +11,8 @@
 from ludwig.modules.metric_registry import get_metric_classes, metric_feature_registry  # noqa: F401
 from ludwig.utils.data_utils import load_json
 
+logger = logging.getLogger(__name__)
+
 
 @dataclass
 class MetricDiff:
@@ -197,7 +199,7 @@ def export_metrics_diff_to_csv(metrics_diff: MetricsDiff, path: str):
                     "Diff Percentage": diff_percentage,
                 }
             )
-            logging.info(f"Exported a CSV report to {path}")
+            logger.info(f"Exported a CSV report to {path}")
 
 
 def build_metrics_summary(experiment_local_directory: str) -> MetricsSummary:
@@ -360,7 +362,7 @@ def export_resource_usage_diff_to_csv(resource_usage_diff: ResourceUsageDiff, pa
                     "Diff Percentage": diff_percentage,
                 }
             )
-            logging.info(f"Exported a CSV report to {path}")
+            logger.info(f"Exported a CSV report to {path}")
 
 
 def average_runs(path_to_runs_dir: str) -> Dict[str, Union[int, float]]:

diff --git a/ludwig/benchmarking/utils.py b/ludwig/benchmarking/utils.py
@@ -18,6 +18,8 @@
 from ludwig.utils.defaults import default_random_seed
 from ludwig.utils.fs_utils import get_fs_and_path
 
+logger = logging.getLogger(__name__)
+
 
 def load_from_module(
     dataset_module: Union[BaseDataset, ModuleType], output_feature: Dict[str, str], subsample_frac: float = 1
@@ -61,7 +63,7 @@ def export_artifacts(experiment: Dict[str, str], experiment_output_directory: st
             os.path.join("configs", experiment["config_path"]),
             os.path.join(export_full_path, CONFIG_YAML),
         )
-        logging.info(f"Uploaded experiment artifact to\n\t{export_full_path}")
+        logger.info(f"Uploaded experiment artifact to\n\t{export_full_path}")
     except Exception:
         logging.exception(
             f"Failed to upload experiment artifacts for experiment *{experiment['experiment_name']}* on "