[ENH] feat :replacing print() statements with Python logging module. (#477)

#### Reference Issues/PRs
Fixes #416.

#### What does this implement/fix? Explain your changes.
This PR migrates the codebase from using `print()` statements for
progress and debug reporting to the standard Python `logging` module.
This allows library consumers to control log verbosity programmatically
and prevents cluttering `stdout` in production environments.

**Key Changes:**
* **Module-level Loggers:** Added `logger = logging.getLogger(__name__)`
to `mcts/_algorithm.py`, `aptatrans/_pipeline.py`, and
`aptatrans/_model.py`.
*   **Logging levels:** 
* Used `logger.debug()` for high-frequency MCTS round updates and
internal search progress.
* Used `logger.info()` for lifecycle events like model weight loading,
downloads, and final candidate recommendations.
* **Performance Optimization:** Implemented
`logger.isEnabledFor(logging.DEBUG)` guards in the MCTS hot loop (`run`
method) to avoid unnecessary string formatting overhead when debug
logging is disabled.
* **Ruff Enforcement:** Updated `pyproject.toml` to include the `T20`
(flake8-print) rule set in the Ruff configuration. This will cause
linting to fail if `print()` or `pprint()` statements are accidentally
reintroduced in the future.

#### What should a reviewer concentrate their feedback on?
* Verify that the chosen log levels (`DEBUG` vs `INFO`) are appropriate
for the respective output.
* Check the implementation of the performance guards in
`pyaptamer/mcts/_algorithm.py`.

#### Did you add any tests for the change?
The changes were verified by running `ruff check pyaptamer/ --select
T20` to ensure that no `print()` statements remain in the library code
and that the new linting rule is active. Syntax correctness was verified
with a full AST parse check of the modified files.

#### PR checklist
- [x] The PR title starts with either [ENH], [MNT], [DOC], or [BUG].
- [ ] Added/modified tests
- [x] Used pre-commit hooks when committing to ensure that code is
compliant with hooks.

Author: ChandanKT-git

pyaptamer/__init__.py

@@ -2,4 +2,8 @@
 
 from importlib.metadata import version
 
+from pyaptamer._logger import logger
+
 __version__ = version("pyaptamer")
+
+__all__ = ["__version__", "logger"]

pyaptamer/_logger.py

@@ -0,0 +1,19 @@
+"""Centralized logger for pyaptamer."""
+
+import logging
+
+# Create the one global logger
+logger = logging.getLogger("pyaptamer")
+logger.setLevel(logging.INFO)
+
+# Format it to show the calling module so we know where logs come from
+formatter = logging.Formatter("[%(levelname)s] [%(module)s] %(message)s")
+
+# Prevent adding multiple handlers if the module is reloaded
+if not logger.handlers:
+    handler = logging.StreamHandler()
+    handler.setFormatter(formatter)
+    logger.addHandler(handler)
+
+# Prevent log propagation to the root logger to avoid duplicate logs
+logger.propagate = False

pyaptamer/aptatrans/_model.py

@@ -11,6 +11,7 @@
 import torch.nn as nn
 from torch import Tensor
 
+from pyaptamer import logger
 from pyaptamer.aptatrans.layers._convolutional import ConvBlock
 from pyaptamer.aptatrans.layers._encoder import (
     EncoderPredictorConfig,
@@ -270,10 +271,10 @@ def load_pretrained_weights(self) -> None:
         )
 
         if os.path.exists(path):
-            print(f"Loading pretrained weights from {path}...")
+            logger.info("Loading pretrained weights from %s...", path)
             state_dict = torch.load(path, map_location=torch.device("cpu"))
         else:
-            print("Downloading best weights from hugging face...")
+            logger.info("Downloading best weights from hugging face...")
             url = (
                 "https://huggingface.co/gcos/pyaptamer-aptatrans/resolve/main/"
                 "pretrained.pt"

pyaptamer/aptatrans/_pipeline.py

@@ -6,9 +6,11 @@
 __author__ = ["nennomp"]
 __all__ = ["AptaTransPipeline"]
 
+
 import torch
 from torch import Tensor
 
+from pyaptamer import logger
 from pyaptamer.aptatrans import AptaTrans
 from pyaptamer.experiments import AptamerEvalAptaTrans
 from pyaptamer.mcts import MCTS
@@ -246,8 +248,11 @@ def recommend(
 
         if verbose:
             for candidate, sequence, score in candidates.values():
-                print(
-                    f"Candidate: {candidate}, Sequence: {sequence}, Score: {score:.4f}"
+                logger.info(
+                    "Candidate: %s, Sequence: %s, Score: %.4f",
+                    candidate,
+                    sequence,
+                    score,
                 )
 
         return set(candidates.values())

pyaptamer/mcts/_algorithm.py

@@ -3,11 +3,14 @@
 __author__ = ["nennomp"]
 __all__ = ["MCTS"]
 
+import logging
 import random
 
 import numpy as np
 from skbase.base import BaseObject
 
+from pyaptamer import logger
+
 
 class MCTS(BaseObject):
     """
@@ -296,8 +299,8 @@ def run(self, verbose: bool = True) -> dict:
         # continue until we reach the target sequence length (i.e, depth * 2)
         round_count = 0
         while len(self.base) < self.depth * 2:
-            if verbose:
-                print(f"\n ----- Round: {round_count + 1} -----")
+            if verbose and logger.isEnabledFor(logging.DEBUG):
+                logger.debug("Round: %d", round_count + 1)
 
             for _ in range(self.n_iterations):
                 # selection
@@ -315,11 +318,12 @@ def run(self, verbose: bool = True) -> dict:
 
             self.base = self._find_best_subsequence()
 
-            if verbose:
-                print("#" * 50)
-                print(f"Best subsequence: {self.base}")
-                print(f"Depth: {len(self.base) // 2}")
-                print("#" * 50)
+            if verbose and logger.isEnabledFor(logging.DEBUG):
+                logger.debug(
+                    "Best subsequence: %s | Depth: %d",
+                    self.base,
+                    len(self.base) // 2,
+                )
 
             # reset for next iteration
             self.root = TreeNode(

pyproject.toml

@@ -39,7 +39,7 @@ target-version = "py310"
 exclude = ["__pycache__", ".venv", "build", "dist", ".git"]
 
 [tool.ruff.lint]
-select = ["E", "F", "B", "I", "UP", "N", "C4"]
+select = ["E", "F", "B", "I", "UP", "N", "C4", "T20"]
 ignore = ["N803", "N806", "N812"]
 
 [tool.ruff.format]
@@ -56,3 +56,6 @@ include-package-data = true
 
 [tool.setuptools.package-data]
 pyaptamer = ["datasets/data/*.csv", "datasets/data/*.pdb"]
+
+[tool.ruff.lint.per-file-ignores]
+"examples/*.ipynb" = ["T201"]