Minimalistic setup.py

gradio_demo.py

@@ -8,7 +8,7 @@
 
 
 import base64, os
-from util.utils import check_ocr_box, get_yolo_model, get_caption_model_processor, get_som_labeled_img
+from omniparser.utils import check_ocr_box, get_yolo_model, get_caption_model_processor, get_som_labeled_img
 import torch
 from PIL import Image
 

util/omniparser.py → omniparser/omniparser.py

@@ -1,4 +1,4 @@
-from util.utils import get_som_labeled_img, get_caption_model_processor, get_yolo_model, check_ocr_box
+from omniparser.utils import get_som_labeled_img, get_caption_model_processor, get_yolo_model, check_ocr_box
 import torch
 from PIL import Image
 import io

util/utils.py → omniparser/utils.py

@@ -41,7 +41,7 @@
 from torchvision.transforms import ToPILImage
 import supervision as sv
 import torchvision.transforms as T
-from util.box_annotator import BoxAnnotator 
+from omniparser.box_annotator import BoxAnnotator 
 
 
 def get_caption_model_processor(model_name, model_name_or_path="Salesforce/blip2-opt-2.7b", device=None):

omnitool/omniparserserver/omniparserserver.py

@@ -11,7 +11,7 @@
 import uvicorn
 root_dir = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 sys.path.append(root_dir)
-from util.omniparser import Omniparser
+from omniparser.omniparser import Omniparser
 
 def parse_arguments():
     parser = argparse.ArgumentParser(description='Omniparser API')

setup.py

@@ -0,0 +1,42 @@
+# This project is licensed under the Creative Commons Attribution 4.0 International License.
+# See https://creativecommons.org/licenses/by/4.0/ for details.
+
+from setuptools import setup, find_namespace_packages
+import pathlib
+
+here = pathlib.Path(__file__).parent.resolve()
+long_description = (here / "README.md").read_text(encoding="utf-8")
+
+setup(
+    name = "omniparser",
+    version = "0.1.0",
+    author = "Microsoft Research AIF Frontiers",
+    description = 'OmniParser, comprehensive method for parsing user interface screenshots into structured and easy-to-understand elements',
+    long_description = long_description,
+    long_description_content_type = "text/markdown",
+    include_package_data = True,
+    packages = find_namespace_packages(where='omniparser'),
+    package_dir = {"omniparser": "omniparser"},
+    install_requires = [
+        "torch",
+        "easyocr",
+        "torchvision",
+        "supervision==0.18.0",
+        "transformers",
+        "ultralytics==8.3.70",
+        "numpy==1.26.4",
+        "opencv-python",
+        "opencv-python-headless",
+        "gradio",
+        "dill",
+        "accelerate",
+        "timm",
+        "einops==0.8.0",
+        "paddlepaddle",
+        "paddleocr",
+    ],
+    classifiers=[
+        "License :: CC-BY-4.0"
+    ],
+    license_files = ['LICENSE']
+)