Add deepmerge dependency

pyproject.toml

@@ -30,6 +30,7 @@ dependencies = [
   "hydra-core>=1.3.2",
   "numba>=0.60",
   "loguru>=0.7.3",
+  "deepmerge>=2.0",
 ]
 requires-python = ">=3.9,<3.13"
 readme = "README.md"

src/batdetect2/api_v2.py

@@ -8,6 +8,7 @@ from soundevent.audio.files import get_audio_files
 
 from batdetect2.audio import build_audio_loader
 from batdetect2.config import BatDetect2Config
+from batdetect2.core import merge_configs
 from batdetect2.evaluate import DEFAULT_EVAL_DIR, build_evaluator, evaluate
 from batdetect2.inference import process_file_list, run_batch_inference
 from batdetect2.logging import DEFAULT_LOGS_DIR
@@ -61,6 +62,7 @@ class BatDetect2API:
         checkpoint_dir: Optional[Path] = DEFAULT_CHECKPOINT_DIR,
         log_dir: Optional[Path] = DEFAULT_LOGS_DIR,
         experiment_name: Optional[str] = None,
+        num_epochs: Optional[int] = None,
         run_name: Optional[str] = None,
         seed: Optional[int] = None,
     ):
@@ -75,6 +77,7 @@ class BatDetect2API:
             val_workers=val_workers,
             checkpoint_dir=checkpoint_dir,
             log_dir=log_dir,
+            num_epochs=num_epochs,
             experiment_name=experiment_name,
             run_name=run_name,
             seed=seed,
@@ -243,7 +246,9 @@ class BatDetect2API:
     ):
         model, stored_config = load_model_from_checkpoint(path)
 
-        config = config or stored_config
+        config = (
+            merge_configs(config, stored_config) if config else stored_config
+        )
 
         targets = build_targets(config=config.targets)
 

src/batdetect2/cli/base.py

@@ -1,5 +1,7 @@
 """BatDetect2 command line interface."""
 
+import sys
+
 import click
 from loguru import logger
 
@@ -19,8 +21,28 @@ BatDetect2 - Detection and Classification
 
 
 @click.group()
-def cli():
+@click.option(
+    "-v",
+    "--verbose",
+    count=True,
+    help="Increase verbosity. -v for INFO, -vv for DEBUG.",
+)
+def cli(
+    verbose: int = 0,
+):
     """BatDetect2 - Bat Call Detection and Classification."""
     click.echo(INFO_STR)
+
+    logger.remove()
+
+    if verbose == 0:
+        log_level = "WARNING"
+    elif verbose == 1:
+        log_level = "INFO"
+    else:
+        log_level = "DEBUG"
+
+    logger.add(sys.stderr, level=log_level)
+
     logger.enable("batdetect2")
     # click.echo(BATDETECT_ASCII_ART)

src/batdetect2/cli/data.py

@@ -35,9 +35,11 @@ def summary(
     from batdetect2.data import load_dataset_from_config
 
     base_dir = base_dir or Path.cwd()
+
     dataset = load_dataset_from_config(
         dataset_config,
         field=field,
         base_dir=base_dir,
     )
+
     print(f"Number of annotated clips: {len(dataset)}")

src/batdetect2/cli/evaluate.py

@@ -1,4 +1,3 @@
-import sys
 from pathlib import Path
 from typing import Optional
 
@@ -22,12 +21,6 @@ DEFAULT_OUTPUT_DIR = Path("outputs") / "evaluation"
 @click.option("--experiment-name", type=str)
 @click.option("--run-name", type=str)
 @click.option("--workers", "num_workers", type=int)
-@click.option(
-    "-v",
-    "--verbose",
-    count=True,
-    help="Increase verbosity. -v for INFO, -vv for DEBUG.",
-)
 def evaluate_command(
     model_path: Path,
     test_dataset: Path,
@@ -37,27 +30,18 @@ def evaluate_command(
     num_workers: Optional[int] = None,
     experiment_name: Optional[str] = None,
     run_name: Optional[str] = None,
-    verbose: int = 0,
 ):
     from batdetect2.api_v2 import BatDetect2API
     from batdetect2.config import load_full_config
     from batdetect2.data import load_dataset_from_config
 
-    logger.remove()
-    if verbose == 0:
-        log_level = "WARNING"
-    elif verbose == 1:
-        log_level = "INFO"
-    else:
-        log_level = "DEBUG"
-    logger.add(sys.stderr, level=log_level)
-
     logger.info("Initiating evaluation process...")
 
     test_annotations = load_dataset_from_config(
         test_dataset,
         base_dir=base_dir,
     )
+
     logger.debug(
         "Loaded {num_annotations} test examples",
         num_annotations=len(test_annotations),

src/batdetect2/cli/train.py

@@ -1,4 +1,3 @@
-import sys
 from pathlib import Path
 from typing import Optional
 
@@ -21,15 +20,10 @@ __all__ = ["train_command"]
 @click.option("--config-field", type=str)
 @click.option("--train-workers", type=int)
 @click.option("--val-workers", type=int)
+@click.option("--num-epochs", type=int)
 @click.option("--experiment-name", type=str)
 @click.option("--run-name", type=str)
 @click.option("--seed", type=int)
-@click.option(
-    "-v",
-    "--verbose",
-    count=True,
-    help="Increase verbosity. -v for INFO, -vv for DEBUG.",
-)
 def train_command(
     train_dataset: Path,
     val_dataset: Optional[Path] = None,
@@ -40,11 +34,11 @@ def train_command(
     targets_config: Optional[Path] = None,
     config_field: Optional[str] = None,
     seed: Optional[int] = None,
+    num_epochs: Optional[int] = None,
     train_workers: int = 0,
     val_workers: int = 0,
     experiment_name: Optional[str] = None,
     run_name: Optional[str] = None,
-    verbose: int = 0,
 ):
     from batdetect2.api_v2 import BatDetect2API
     from batdetect2.config import (
@@ -54,14 +48,6 @@ def train_command(
     from batdetect2.data import load_dataset_from_config
     from batdetect2.targets import load_target_config
 
-    logger.remove()
-    if verbose == 0:
-        log_level = "WARNING"
-    elif verbose == 1:
-        log_level = "INFO"
-    else:
-        log_level = "DEBUG"
-    logger.add(sys.stderr, level=log_level)
     logger.info("Initiating training process...")
 
     logger.info("Loading configuration...")
@@ -99,7 +85,10 @@ def train_command(
     if model_path is None:
         api = BatDetect2API.from_config(conf)
     else:
-        api = BatDetect2API.from_checkpoint(model_path)
+        api = BatDetect2API.from_checkpoint(
+            model_path,
+            config=conf if config is not None else None,
+        )
 
     return api.train(
         train_annotations=train_annotations,
@@ -108,6 +97,7 @@ def train_command(
         val_workers=val_workers,
         checkpoint_dir=ckpt_dir,
         log_dir=log_dir,
+        num_epochs=num_epochs,
         experiment_name=experiment_name,
         run_name=run_name,
         seed=seed,

src/batdetect2/core/__init__.py

@@ -1,8 +1,9 @@
-from batdetect2.core.configs import BaseConfig, load_config
+from batdetect2.core.configs import BaseConfig, load_config, merge_configs
 from batdetect2.core.registries import Registry
 
 __all__ = [
     "BaseConfig",
     "load_config",
     "Registry",
+    "merge_configs",
 ]

src/batdetect2/core/configs.py

@@ -11,12 +11,14 @@ configuration sections.
 from typing import Any, Optional, Type, TypeVar
 
 import yaml
+from deepmerge.merger import Merger
 from pydantic import BaseModel, ConfigDict
 from soundevent.data import PathLike
 
 __all__ = [
     "BaseConfig",
     "load_config",
+    "merge_configs",
 ]
 
 
@@ -178,3 +180,19 @@ def load_config(
         config = get_object_field(config, field)
 
     return schema.model_validate(config or {})
+
+
+default_merger = Merger(
+    [],
+    ["override"],
+    ["override"],
+)
+
+
+def merge_configs(config1: T, config2: T) -> T:
+    """Merge two configuration objects."""
+    model = type(config1)
+    dict1 = config1.model_dump()
+    dict2 = config2.model_dump()
+    merged = default_merger.merge(dict1, dict2)
+    return model.model_validate(merged)

src/batdetect2/train/train.py

@@ -47,6 +47,7 @@ def train(
     checkpoint_dir: Optional[Path] = None,
     log_dir: Optional[Path] = None,
     experiment_name: Optional[str] = None,
+    num_epochs: Optional[int] = None,
     run_name: Optional[str] = None,
     seed: Optional[int] = None,
 ):
@@ -107,6 +108,7 @@ def train(
             targets=targets,
         ),
         checkpoint_dir=checkpoint_dir,
+        num_epochs=num_epochs,
         log_dir=log_dir,
         experiment_name=experiment_name,
         run_name=run_name,
@@ -128,6 +130,7 @@ def build_trainer(
     log_dir: Optional[Path] = None,
     experiment_name: Optional[str] = None,
     run_name: Optional[str] = None,
+    num_epochs: Optional[int] = None,
 ) -> Trainer:
     trainer_conf = config.train.trainer
     logger.opt(lazy=True).debug(
@@ -149,8 +152,13 @@ def build_trainer(
         )
     )
 
+    train_config = trainer_conf.model_dump(exclude_none=True)
+
+    if num_epochs is not None:
+        train_config["max_epochs"] = num_epochs
+
     return Trainer(
-        **trainer_conf.model_dump(exclude_none=True),
+        **train_config,
         logger=train_logger,
         callbacks=[
             build_checkpoint_callback(