lecrapaud 0.16.3__tar.gz → 0.16.5__tar.gz

{lecrapaud-0.16.3 → lecrapaud-0.16.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: lecrapaud
-Version: 0.16.3
+Version: 0.16.5
 Summary: Framework for machine and deep learning, with regression, classification and time series analysis
 License: Apache License
 Author: Pierre H. Gallet

{lecrapaud-0.16.3 → lecrapaud-0.16.5}/lecrapaud/config.py

@@ -25,7 +25,7 @@ DB_PORT = (
 DB_NAME = (
     os.getenv("TEST_DB_NAME") if PYTHON_ENV == "Test" else os.getenv("DB_NAME", None)
 )
-DB_URI = (
+DB_URI: str = (
     os.getenv("TEST_DB_URI", None)
     if PYTHON_ENV == "Test"
     else os.getenv("DB_URI", None)

{lecrapaud-0.16.3 → lecrapaud-0.16.5}/lecrapaud/db/alembic.ini

@@ -82,7 +82,7 @@ sqlalchemy.url = %(DATABASE_URL)s
 
 # Logging configuration
 [loggers]
-keys = root,sqlalchemy,alembic
+keys = root,sqlalchemy,alembic,lecrapaud
 
 [loggers_root]
 disable_existing_loggers = False
@@ -108,6 +108,12 @@ level = INFO
 handlers =
 qualname = alembic
 
+[logger_lecrapaud]
+level = INFO
+handlers = console
+qualname = lecrapaud
+propagate = 0
+
 [handler_console]
 class = StreamHandler
 args = (sys.stderr,)

{lecrapaud-0.16.3 → lecrapaud-0.16.5}/lecrapaud/db/models/experiment.py

@@ -241,7 +241,7 @@ class Experiment(Base):
             # This ensures we're comparing apples to apples by normalizing the scores
             experiments = db.query(cls).filter(cls.name.ilike(f"%{name}%")).all()
             if not experiments:
-                return None
+                raise ValueError(f"No experiments found with the given name: {name}")
 
             # Get all scores
             rmse_scores = [e.avg_rmse for e in experiments if e.avg_rmse is not None]
@@ -250,7 +250,9 @@ class Experiment(Base):
             ]
 
             if not rmse_scores or not logloss_scores:
-                return None
+                raise ValueError(
+                    "No experiments found with both RMSE and LogLoss scores. Maybe try with only one metric."
+                )
 
             # Normalize scores (subtract min and divide by range)
             min_rmse = min(rmse_scores)
@@ -277,17 +279,29 @@ class Experiment(Base):
 
             return experiment_scores[0][0] if experiment_scores else None
 
-        # For single metric case (rmse or logloss)
-        score_property = cls.avg_rmse if metric == "rmse" else cls.avg_logloss
+        elif metric == "rmse" or metric == "logloss":
+            # For single metric case (rmse or logloss)
+            # Need to get all experiments first to evaluate instance properties
+            experiments = db.query(cls).filter(cls.name.ilike(f"%{name}%")).all()
 
-        return (
-            db.query(cls)
-            .filter(
-                cls.name.ilike(f"%{name}%"), score_property.isnot(None)
-            )  # Only consider experiments with scores
-            .order_by(score_property)
-            .first()
-        )
+            if not experiments:
+                return None
+
+            # Filter out experiments without scores and sort by the selected metric
+            filtered_experiments = []
+            for exp in experiments:
+                score = exp.avg_rmse if metric == "rmse" else exp.avg_logloss
+                if score is not None:
+                    filtered_experiments.append((exp, score))
+
+            if not filtered_experiments:
+                return None
+
+            # Sort by score (ascending since lower is better)
+            filtered_experiments.sort(key=lambda x: x[1])
+            return filtered_experiments[0][0]
+        else:
+            raise ValueError("Invalid metric. Must be 'rmse', 'logloss', or 'both'.")
 
     def get_features(self, target_number: int):
         targets = [t for t in self.targets if t.name == f"TARGET_{target_number}"]

{lecrapaud-0.16.3 → lecrapaud-0.16.5}/lecrapaud/db/session.py

@@ -13,6 +13,8 @@ from lecrapaud.config import DB_USER, DB_PASSWORD, DB_HOST, DB_PORT, DB_NAME, DB
 _engine = None
 _SessionLocal = None
 if DB_URI:
+    if "mysql://" in DB_URI:
+        DB_URI = DB_URI.replace("mysql://", "mysql+pymysql://")
     DATABASE_URL = DB_URI
 elif DB_USER:
     DATABASE_URL = (
@@ -23,15 +25,13 @@ else:
 
 
 def init_db(uri: str = None):
-    global _engine, _SessionLocal, DATABASE_URL
+    global _engine, _SessionLocal, DATABASE_URL, DB_URI
     if uri:
+        if "mysql://" in uri:
+            uri = uri.replace("mysql://", "mysql+pymysql://")
         DATABASE_URL = uri
-    elif DB_USER:
-        DATABASE_URL = (
-            f"mysql+pymysql://{DB_USER}:{DB_PASSWORD}@{DB_HOST}:{DB_PORT}/{DB_NAME}"
-        )
-    elif DB_URI:
-        DATABASE_URL = DB_URI
+    elif DATABASE_URL:
+        pass
     else:
         raise ValueError(
             "No database configuration found, please set env variables "

{lecrapaud-0.16.3 → lecrapaud-0.16.5}/lecrapaud/model_selection.py

@@ -1253,7 +1253,7 @@ class ModelSelectionEngine:
                 try:
                     with open(best_params_file, "r") as f:
                         target_params = json.load(f)
-                        all_best_params[target_number] = target_params
+                        all_best_params[int(target_number)] = target_params
                 except Exception as e:
                     logger.warning(
                         f"Error loading best params for {target_dir}: {str(e)}"

{lecrapaud-0.16.3 → lecrapaud-0.16.5}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "lecrapaud"
-version = "0.16.3"
+version = "0.16.5"
 description = "Framework for machine and deep learning, with regression, classification and time series analysis"
 authors = [
     {name = "Pierre H. Gallet"}