update config

README.md

@@ -41,9 +41,9 @@ comic-panel-extractor path/to/comic.jpg
 **Python:**
 ```python
 from comic_panel_extractor.main import ComicPanelExtractor
-from comic_panel_extractor.config import Config
+from comic_panel_extractor.config import Config, load_config
 
-config = Config()
+config = load_config()
 config.input_path = "comic.jpg"
 extractor = ComicPanelExtractor(config)
 panels, data, paths = extractor.extract_panels_from_comic()
@@ -99,7 +99,7 @@ with YOLOManager() as yolo:
 ## Configuration
 
 ```python
-config = Config()
+config = load_config()
 config.min_width_ratio = 0.15    # Min panel width (% of image)
 config.min_height_ratio = 0.15   # Min panel height (% of image)
 config.min_area_ratio = 0.05     # Min panel area (% of image)

comic_panel_extractor/annorator_server.py

@@ -18,7 +18,7 @@ import psutil
 import subprocess
 from . import common
 import fcntl
-from .config import load_config, update_toml_key
+from .config import Config, load_config, update_toml_key
 
 app = APIRouter()
 
@@ -423,7 +423,16 @@ async def save_config(request: TrainConfig):
     update_toml_key("RESUME_TRAIN", request.resume_train)
 
     return {'message': 'Config update successfully.', 'status': 'success'}
-    
+
+@app.post("/api/annotate/model_reset")
+async def reset_model():
+    from pathlib import Path
+    file_path = Path(config.yolo_trained_model_path)
+
+    if file_path.exists():
+        file_path.unlink()
+
+    return {'message': 'Model Reseted', 'status': 'success'}
 
 @app.get("/api/annotate/train")
 async def upload_image():

comic_panel_extractor/border_panel_extractor.py

@@ -10,7 +10,7 @@ from scipy import ndimage as ndi
 from skimage.morphology import remove_small_holes
 import cv2
 
-from .config import Config
+from .config import Config, load_config
 from .image_processor import ImageProcessor
 from .utils import remove_duplicate_boxes, count_panels_inside, extend_boxes_to_image_border
 
@@ -27,7 +27,7 @@ class BorderPanelExtractor:
     
     def __init__(self, config: Config = None):
         """Initialize the BorderPanelExtractor with optional configuration."""
-        self.config = config or Config()
+        self.config = config or load_config()
         self.output_folder = f'{self.config.output_folder}/border_panel_extractor'
         os.makedirs(self.output_folder, exist_ok=True)
         self.PANEL_FILENAME_PATTERN = re.compile(self.config.panel_filename_pattern)
@@ -640,7 +640,7 @@ class BorderPanelExtractor:
 
 
 if __name__ == "__main__":
-    config = Config()
+    config = load_config()
     config.input_path = "test0.jpg"
 
     import shutil

comic_panel_extractor/cli.py

@@ -9,7 +9,7 @@ import json
 from typing import Optional, List
 
 from .main import ComicPanelExtractor
-from .config import Config
+from .config import Config, load_config
 
 
 class ComicPanelCLI:
@@ -58,7 +58,7 @@ Examples:
 	
 	def _load_config(self, args: argparse.Namespace) -> Config:
 		"""Load configuration from file or create from arguments."""
-		config = Config()
+		config = load_config()
 		
 		# Load from config file if provided
 		if args.config:

comic_panel_extractor/config.py

@@ -28,6 +28,7 @@ class Config:
     YOLO_BASE_MODEL_NAME: str = "yolo11s-seg"
     YOLO_MODEL_NAME: str = ""  # will be derived if empty
     IMAGE_SOURCE_PATH: str = ""
+    YOLO_MODEL_REMOTE_URL: str = ""
 
     # Derived paths
     yolo_base_model_path: str = ""
@@ -88,7 +89,8 @@ def load_config(file_path=CONFIG_FILE) -> Config:
         RECREATE_DATASET=to_bool(data.get("RECREATE_DATASET", True)),
         YOLO_BASE_MODEL_NAME=data.get("YOLO_BASE_MODEL_NAME", "yolo11s-seg"),
         YOLO_MODEL_NAME=data.get("YOLO_MODEL_NAME", ""),  # derived in __post_init__
-        IMAGE_SOURCE_PATH=data.get("IMAGE_SOURCE_PATH", "")
+        IMAGE_SOURCE_PATH=data.get("IMAGE_SOURCE_PATH", ""),
+        YOLO_MODEL_REMOTE_URL=data.get("YOLO_MODEL_REMOTE_URL", "")
     )
 
 

comic_panel_extractor/config.toml

@@ -1,4 +1,4 @@
-EPOCH = 200
+EPOCH = 500
 DEFAULT_IMAGE_SIZE = 640
 BATCH = 10
 RESUME_TRAIN = true
@@ -6,3 +6,4 @@ RECREATE_DATASET = true
 YOLO_BASE_MODEL_NAME = "yolo11s-seg"
 YOLO_MODEL_NAME = "comic_panel_yolo11s-seg"
 IMAGE_SOURCE_PATH = "images"
+YOLO_MODEL_REMOTE_URL = "https://huggingface.co/mosesb/best-comic-panel-detection/resolve/main/best.pt"

comic_panel_extractor/create_dataset.py

@@ -4,7 +4,7 @@ import random
 from pathlib import Path
 from dotenv import load_dotenv
 from tqdm import tqdm
-from .config import load_config
+from .config import Config, load_config
 
 load_dotenv()
 config = load_config()

comic_panel_extractor/extractor_server.py

@@ -1,7 +1,7 @@
 from fastapi import APIRouter, File, UploadFile, HTTPException
 from fastapi.responses import FileResponse
 import os
-from .config import load_config
+from .config import Config, load_config
 from .main import ComicPanelExtractor
 import traceback
 from pathlib import Path
@@ -76,7 +76,6 @@ async def convert_comic(file: UploadFile = File(...)):
         print("====================================")
 
         # Extract panels
-        config = Config()
         config.input_path = file_path
         config.output_folder = specific_output_folder
 

comic_panel_extractor/image_processor.py

@@ -1,6 +1,6 @@
 from typing import List, Tuple
 from pathlib import Path
-from .config import Config
+from .config import Config, load_config
 
 import numpy as np
 import cv2
@@ -18,7 +18,7 @@ class ImageProcessor:
     """Handles image preprocessing operations."""
     
     def __init__(self, config: Config = None):
-        self.config = config or Config()
+        self.config = config or load_config()
         self.index = 0
 
     def get_output_path(self, output_folder, file_name):

comic_panel_extractor/inference.py

@@ -2,7 +2,7 @@
 from .yolo_manager import YOLOManager
 from .utils import get_abs_path, get_image_paths
 import os
-from .config import load_config
+from .config import Config, load_config
 
 config = load_config()
 

comic_panel_extractor/llm_panel_extractor.py

@@ -1,4 +1,4 @@
-from .config import Config
+from .config import Config, load_config
 from ultralytics import YOLO 
 from PIL import Image 
 import cv2
@@ -14,13 +14,12 @@ class LLMPanelExtractor:
 	"""Handles image preprocessing operations."""
 	
 	def __init__(self, config: Config = None):
-		self.config = config or Config()
+		self.config = config or load_config()
 
 		# Check if YOLO model exists; if not, download it to the specified path
-		yolo_base_model_path = f'{self.config.yolo_base_model_path}_best.pt'
-		# yolo_base_model_path = f'{self.config.yolo_trained_model_path}'
+		yolo_base_model_path = self.config.yolo_trained_model_path
 		if not os.path.exists(yolo_base_model_path):
-			url = "https://huggingface.co/mosesb/best-comic-panel-detection/resolve/main/best.pt"
+			url = self.config.YOLO_MODEL_REMOTE_URL
 			print(f"Downloading YOLO model to {yolo_base_model_path}...")
 			response = requests.get(url)
 			response.raise_for_status()  # Raise an error if the download fails
@@ -158,7 +157,7 @@ class LLMPanelExtractor:
 def extract_panel_via_llm(input_image_path, config=None, reset=True):
 	"""Main function to extract panels using various image processing techniques."""
 	# Initialize configuration
-	extractor_config = config or Config()
+	extractor_config = config or load_config()
 	extractor_config.org_input_path = input_image_path
 
 	# Clean output folder

comic_panel_extractor/main.py

@@ -1,5 +1,5 @@
 # from .text_detector import TextDetector
-from .config import Config
+from .config import Config, load_config
 from .image_processor import ImageProcessor
 from .panel_extractor import PanelData
 from .panel_extractor import PanelExtractor

comic_panel_extractor/panel_extractor.py

@@ -1,5 +1,5 @@
 from typing import List, Tuple
-from .config import Config
+from .config import Config, load_config
 
 import numpy as np
 import cv2

comic_panel_extractor/server.py

@@ -4,7 +4,7 @@ from fastapi.middleware.cors import CORSMiddleware
 from .extractor_server import app as extractor_app, delete_folder_if_old_or_empty, output_folder
 from .annorator_server import app as annotator_app
 import os, json
-from .config import load_config
+from .config import Config, load_config
 
 from fastapi import Request
 from fastapi.responses import HTMLResponse

comic_panel_extractor/static/annotator.html

@@ -1211,7 +1211,6 @@
                 document.getElementById('resetModalBtn').addEventListener('click', async () => {
                     if (confirm('Are you sure you want to reset the model? This action cannot be undone.')) {
                         try {
-                            this.openXterm();
                             const response = await fetch('/api/annotate/model_reset', { method: 'POST' });
                             if (!response.ok) {
                                 throw new Error(`Server error: ${response.status}`);

comic_panel_extractor/text_detector.py

@@ -5,7 +5,7 @@ from typing import List, Optional
 from dataclasses import dataclass
 import numpy as np
 
-from .config import Config, get_text_cood_file_path
+from .config import Config, load_config, get_text_cood_file_path
 
 @dataclass
 class TextDetection:

comic_panel_extractor/train.py

@@ -2,7 +2,7 @@
 from .yolo_manager import YOLOManager
 from .utils import get_abs_path, backup_file
 import os
-from .config import load_config
+from .config import Config, load_config
 import yaml
 import os
 from pathlib import Path

comic_panel_extractor/utils.py

@@ -7,7 +7,7 @@ import os
 import shutil
 from glob import glob
 from typing import List, Union
-from .config import load_config
+from .config import Config, load_config
 from shapely.geometry import Polygon
 
 config = load_config()

comic_panel_extractor/yolo_manager.py

@@ -4,7 +4,7 @@ import shutil
 from glob import glob
 from typing import List, Union
 from . import utils
-from .config import load_config
+from .config import Config, load_config
 config = load_config()
 
 os.environ["TORCH_USE_CUDA_DSA"] = "1"
@@ -56,7 +56,6 @@ import cv2
 from ultralytics import YOLO
 from typing import List, Optional, Dict, Any
 from .utils import get_abs_path, clean_directory
-from .config import Config
 from dotenv import load_dotenv
 load_dotenv()