refactor: extract helper functions in Model.ipynb

RenAIssance_CRNN_OCR_Shashank_Shekhar_Singh/Dataset_Generation.ipynb

@@ -9,7 +9,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
@@ -177,6 +177,18 @@
     "from tensorflow.keras.utils import plot_model"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Helper function to create a directory if it does not already exist\n",
+    "def ensure_dir(path):\n",
+    "    if not os.path.exists(path):\n",
+    "        os.makedirs(path)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {
@@ -284,15 +296,13 @@
     "\n",
     "pdf_path1 = \"./Padilla - Nobleza virtuosa_testExtract.pdf\"  # Path to the PDF file\n",
     "unproc_images_folder_1 = \"./preprocessing/imgsUnProcessed1\"  # Output folder to save the images\n",
-    "if not os.path.exists(unproc_images_folder_1):\n",
-    "    os.makedirs(unproc_images_folder_1)\n",
+    "ensure_dir(unproc_images_folder_1)\n",
     "pdf_to_images(pdf_path1, unproc_images_folder_1)\n",
     "\n",
     "\n",
     "pdf_path2 = \"./Padilla - 2 Noble perfecto_Extract.pdf\"  # Path to the PDF file\n",
     "unproc_images_folder_2 = \"./preprocessing/imgsUnProcessed2\"  # Output folder to save the images\n",
-    "if not os.path.exists(unproc_images_folder_2):\n",
-    "    os.makedirs(unproc_images_folder_2)\n",
+    "ensure_dir(unproc_images_folder_2)\n",
     "pdf_to_images(pdf_path2, unproc_images_folder_2)"
    ]
   },
@@ -339,15 +349,13 @@
     "\n",
     "unproc_images_folder_1 = \"./preprocessing/imgsUnProcessed1\"\n",
     "proc_images_folder_1 = \"./preprocessing/imgsForAllPages1\"\n",
-    "if not os.path.exists(proc_images_folder_1):\n",
-    "    os.makedirs(proc_images_folder_1)\n",
+    "ensure_dir(proc_images_folder_1)\n",
     "process_images(unproc_images_folder_1, proc_images_folder_1)\n",
     "\n",
     "\n",
     "unproc_images_folder_2 = \"./preprocessing/imgsUnProcessed2\"\n",
     "proc_images_folder_2 = \"./preprocessing/imgsForAllPages2\"\n",
-    "if not os.path.exists(proc_images_folder_2):\n",
-    "    os.makedirs(proc_images_folder_2)\n",
+    "ensure_dir(proc_images_folder_2)\n",
     "process_images(unproc_images_folder_2, proc_images_folder_2)\n",
     "\n",
     "print(\"Image processing complete!\")"
@@ -459,14 +467,12 @@
     "\n",
     "bound_box_applied1 = './preprocessing/BoundBoxApplied1/'\n",
     "bound_box_sorted1 = \"./preprocessing/BoundBoxSorted1\"\n",
-    "if not os.path.exists(bound_box_sorted1):\n",
-    "    os.makedirs(bound_box_sorted1)\n",
+    "ensure_dir(bound_box_sorted1)\n",
     "sort_bounding_boxes(bound_box_applied1, bound_box_sorted1)\n",
     "\n",
     "bound_box_applied2 = './preprocessing/BoundBoxApplied2/'\n",
     "bound_box_sorted2 = \"./preprocessing/BoundBoxSorted2\"\n",
-    "if not os.path.exists(bound_box_sorted2):\n",
-    "    os.makedirs(bound_box_sorted2)\n",
+    "ensure_dir(bound_box_sorted2)\n",
     "sort_bounding_boxes(bound_box_applied2, bound_box_sorted2)"
    ]
   },
@@ -546,17 +552,15 @@
     "bound_box_sorted1 = './preprocessing/BoundBoxSorted1'\n",
     "proc_grnd_truth1 = \"./preprocessing/textSplitted1\"\n",
     "TEST_SIZE=6\n",
-    "if not os.path.exists(proc_grnd_truth1):\n",
-    "    os.makedirs(proc_grnd_truth1)\n",
+    "ensure_dir(proc_grnd_truth1)\n",
     "process_textfiles(grnd_truth1, bound_box_sorted1, proc_grnd_truth1, TEST_SIZE)\n",
     "\n",
     "\n",
     "grnd_truth2 = \"./preprocessing/all_text2.txt\" \n",
     "bound_box_sorted2 = './preprocessing/BoundBoxSorted2'\n",
     "proc_grnd_truth2 = \"./preprocessing/textSplitted2\"\n",
     "TEST_SIZE=0\n",
-    "if not os.path.exists(proc_grnd_truth2):\n",
-    "    os.makedirs(proc_grnd_truth2)\n",
+    "ensure_dir(proc_grnd_truth2)\n",
     "process_textfiles(grnd_truth2, bound_box_sorted2, proc_grnd_truth2, TEST_SIZE)\n",
     "print(\"Text splitting complete!\")"
    ]
@@ -598,24 +602,22 @@
     "proc_images_folder_1 = './preprocessing/imgsForAllPages1'\n",
     "bound_box_sorted1 = './preprocessing/BoundBoxSorted1'\n",
     "proc_grnd_truth1 = './preprocessing/textSplitted1'\n",
-    "training_data1 = './traning_data1'\n",
+    "training_data1 = './training_data1'\n",
     "test_size=6\n",
     "train_size = count_files_in_folder(proc_images_folder_1, ['.png', '.jpeg', '.jpg'])- test_size\n",
     "print(\"Training pages \" +  str(train_size))\n",
-    "if not os.path.exists(training_data1):\n",
-    "    os.makedirs(training_data1)\n",
+    "ensure_dir(training_data1)\n",
     "apply_extraction_to_folder_for_train(proc_images_folder_1, bound_box_sorted1, proc_grnd_truth1, training_data1, train_size)\n",
     "\n",
     "\n",
     "proc_images_folder_2 = './preprocessing/imgsForAllPages2'\n",
     "bound_box_sorted2 = './preprocessing/BoundBoxSorted2'\n",
     "proc_grnd_truth2 = './preprocessing/textSplitted2'\n",
-    "training_data2 = './traning_data2'\n",
+    "training_data2 = './training_data2'\n",
     "test_size = 0\n",
-    "train_size = count_files_in_folder(proc_images_folder_1, ['.png', '.jpeg', '.jpg'])- test_size\n",
+    "train_size = count_files_in_folder(proc_images_folder_2, ['.png', '.jpeg', '.jpg'])- test_size\n",
     "print(\"Training pages \" +  str(train_size))\n",
-    "if not os.path.exists(training_data2):\n",
-    "    os.makedirs(training_data2)\n",
+    "ensure_dir(training_data2)\n",
     "apply_extraction_to_folder_for_train(proc_images_folder_2, bound_box_sorted2, proc_grnd_truth2, training_data2, train_size) # better to send no. of pages given in transcription"
    ]
   }