Merge pull request #3 from sinanerdinc/feature/log-management

A logger file has been added for log management.

tasnif/calculations.py

@@ -1,16 +1,8 @@
-import logging
-
 import numpy as np
 from img2vec_pytorch import Img2Vec
-from rich.logging import RichHandler
 from scipy.cluster.vq import kmeans2
 from sklearn.decomposition import PCA
-
-# Configure logging
-LOG_FORMAT = "%(asctime)s - %(levelname)s - %(message)s"
-logging.basicConfig(
-    level="INFO", format=LOG_FORMAT, datefmt="[%X]", handlers=[RichHandler()]
-)
+from .logger import info
 
 
 def get_embeddings(use_gpu=False, images=None):
@@ -19,7 +11,7 @@ def get_embeddings(use_gpu=False, images=None):
     image embeddings.
     """
 
-    logging.info(f"Img2Vec is running on {'GPU' if use_gpu else 'CPU'}...")
+    info(f"Img2Vec is running on {'GPU' if use_gpu else 'CPU'}...")
     img2vec = Img2Vec(cuda=use_gpu)
 
     embeddings = img2vec.get_vec(images, tensor=False)
@@ -44,7 +36,7 @@ def calculate_pca(embeddings, pca_dim):
     n_samples, _ = embeddings.shape
     if n_samples < pca_dim:
         n_components = min(n_samples, pca_dim)
-        logging.info(
+        info(
             f"Number of samples is less than the desired dimension. Setting n_components to {n_components}"
         )
 
@@ -53,7 +45,7 @@ def calculate_pca(embeddings, pca_dim):
 
     pca = PCA(n_components=n_components)
     pca_embeddings = pca.fit_transform(embeddings.squeeze())
-    logging.info("PCA calculated.")
+    info("PCA calculated.")
     return pca_embeddings
 
 
@@ -74,7 +66,7 @@ def calculate_kmeans(pca_embeddings, num_classes):
     try:
         centroid, labels = kmeans2(data=pca_embeddings, k=num_classes, minit="points")
         counts = np.bincount(labels)
-        logging.info("KMeans calculated.")
+        info("KMeans calculated.")
         return centroid, labels, counts
 
     except Exception as e:

tasnif/logger.py

@@ -0,0 +1,20 @@
+import logging
+from rich.logging import RichHandler
+
+log_format = "%(asctime)s - %(levelname)s - %(message)s"
+logging.basicConfig(
+    level="INFO", format=log_format, datefmt="[%X]", handlers=[RichHandler(show_time=False, show_level=False)]
+)
+
+
+def info(msg):
+    logging.info(msg)
+
+
+def error(msg):
+    logging.error(msg)
+
+
+if __name__ == '__main__':
+    info("info message")
+    error("error message")

tasnif/tasnif.py

@@ -1,31 +1,21 @@
-import logging
 import os
 import shutil
 import warnings
 from itertools import compress
 import numpy as np
-
-from rich.logging import RichHandler
 from tqdm import tqdm
-
 from .calculations import calculate_kmeans, calculate_pca, get_embeddings
 from .utils import (
     create_dir,
     create_image_grid,
     read_images_from_directory,
     read_with_pil,
 )
+from .logger import info, error
 
 warnings.filterwarnings("ignore")
 
 
-# Configure logging
-LOG_FORMAT = "%(asctime)s - %(levelname)s - %(message)s"
-logging.basicConfig(
-    level="INFO", format=LOG_FORMAT, datefmt="[%X]", handlers=[RichHandler()]
-)
-
-
 class Tasnif:
     def __init__(self, num_classes, pca_dim=16, use_gpu=False):
         self.num_classes = num_classes
@@ -93,16 +83,13 @@ def export(self, output_folder="./"):
                 try:
                     shutil.copy2(img_path, target_directory)
                 except Exception as e:
-                    logging.error(
-                        f"Error copying {img_path} to {target_directory}: {e}"
-                    )
+                    error(f"Error copying {img_path} to {target_directory}: {e}")
 
             # Create an image grid for the current label
             label_images = list(compress(self.images, label_mask))
             create_image_grid(label_images, project_path, label_number)
 
-        logging.info(f"Exported images and grids to {project_path}")
-
+        info(f"Exported images and grids to {project_path}")
 
     def export_embeddings(self, output_folder="./"):
         """
@@ -111,11 +98,10 @@ def export_embeddings(self, output_folder="./"):
         Parameters:
         - output_folder (str): The directory to export the embeddings into.
         """
-        
+
         if self.embeddings is None:
             raise ValueError("Embeddings can not be empty. Please call the calculate method first.")
-
-
+
         embeddings_path = os.path.join(output_folder, f"{self.project_name}_embeddings.npy")
         np.save(embeddings_path, self.embeddings)
-        logging.info(f"Embeddings have been saved to {embeddings_path}")
+        info(f"Embeddings have been saved to {embeddings_path}")

tasnif/utils.py

@@ -1,18 +1,10 @@
 import glob
-import logging
 import os
 from pathlib import Path
-
 import matplotlib.pyplot as plt
 from PIL import Image
-from rich.logging import RichHandler
 from tqdm import tqdm
-
-# Configure logging
-LOG_FORMAT = "%(asctime)s - %(levelname)s - %(message)s"
-logging.basicConfig(
-    level="INFO", format=LOG_FORMAT, datefmt="[%X]", handlers=[RichHandler()]
-)
+from .logger import info
 
 
 def read_images_from_directory(image_directory: str) -> list:
@@ -25,7 +17,7 @@ def read_images_from_directory(image_directory: str) -> list:
     image_extensions = ("*.gif", "*.png", "*.jpg", "*.jpeg")
     for ext in image_extensions:
         list_of_images.extend(glob.glob(os.path.join(image_directory, ext)))
-    logging.info(f"Images found: {len(list_of_images)}")
+    info(f"Images found: {len(list_of_images)}")
     return list_of_images
 
 
@@ -42,7 +34,7 @@ def read_with_pil(list_of_images: list, resize=True) -> list:
         if resize:
             img.thumbnail((512, 512))
         pil_images.append(img)
-    logging.info("Image reading done!")
+    info("Image reading done!")
     return pil_images