python310Packages.paddleocr: init at 2.7.0.1

{ lib

, buildPythonPackage

, pythonRelaxDepsHook

, fetchFromGitHub

, attrdict

, beautifulsoup4

, cython

, fire

, fonttools

, lmdb

, lxml

, numpy

, opencv4

, openpyxl

, pdf2docx

, pillow

, premailer

, pyclipper

, pymupdf

, python-docx

, rapidfuzz

, scikit-image

, shapely

, tqdm

, paddlepaddle

, lanms-neo

, polygon3

}:

let

  version = "2.7.0.1";

in

buildPythonPackage {

  pname = "paddleocr";

  inherit version;

  format = "setuptools";

  src = fetchFromGitHub {

    owner = "PaddlePaddle";

    repo = "PaddleOCR";

    rev = "254786752a2659e184822b4b2de5637a05236590";

    hash = "sha256-M/Fpk9swX9Gds7o5poM9Iv6LOhKoZNbe0Wv9JNMPOU0=";

  };

  patches = [

    # The `ppocr.data.imaug` re-exports the `IaaAugment` and `CopyPaste`

    # classes. These classes depend on the `imgaug` package which is

    # unmaintained and has been removed from nixpkgs.

    #

    # The image OCR feature of PaddleOCR doesn't use these classes though, so

    # they work even after stripping the the `IaaAugment` and `CopyPaste`

    # exports. It probably breaks some of the OCR model creation tooling that

    # PaddleOCR provides, however.

    ./remove-import-imaug.patch

  ];

  nativeBuildInputs = [ pythonRelaxDepsHook ];

  # trying to relax only pymupdf makes the whole build fail

  pythonRelaxDeps = true;

  pythonRemoveDeps = [

    "imgaug"

    "visualdl"

    "opencv-python"

    "opencv-contrib-python"

  ];

  propagatedBuildInputs = [

    attrdict

    beautifulsoup4

    cython

    fire

    fonttools

    lmdb

    lxml

    numpy

    opencv4

    openpyxl

    pdf2docx

    pillow

    premailer

    pyclipper

    pymupdf

    python-docx

    rapidfuzz

    scikit-image

    shapely

    tqdm

    paddlepaddle

    lanms-neo

    polygon3

  ];

  # TODO: The tests depend, among possibly other things, on `cudatoolkit`.

  # But Cudatoolkit fails to install.

  # preCheck = "export HOME=$TMPDIR";

  # nativeCheckInputs = with pkgs; [ which cudatoolkit ];

  doCheck = false;

  meta = with lib; {

    homepage = "https://github.com/PaddlePaddle/PaddleOCR";

    license = licenses.asl20;

    description = "Multilingual OCR toolkits based on PaddlePaddle";

    longDescription = ''

      PaddleOCR aims to create multilingual, awesome, leading, and practical OCR

      tools that help users train better models and apply them into practice.

    '';

    changelog = "https://github.com/PaddlePaddle/PaddleOCR/releases/tag/v${version}";

    maintainers = with maintainers; [ happysalada ];

    platforms = [ "x86_64-linux" "x86_64-darwin" "aarch64-darwin" ];

  };

}

diff --git a/ppocr/data/imaug/__init__.py b/ppocr/data/imaug/__init__.py

index 121582b4..a6987c75 100644

--- a/ppocr/data/imaug/__init__.py

+++ b/ppocr/data/imaug/__init__.py

@@ -16,7 +16,6 @@ from __future__ import division

 from __future__ import print_function

 from __future__ import unicode_literals

-from .iaa_augment import IaaAugment

 from .make_border_map import MakeBorderMap

 from .make_shrink_map import MakeShrinkMap

 from .random_crop_data import EastRandomCropData, RandomCropImgMask

@@ -30,7 +29,6 @@ from .rec_img_aug import BaseDataAugmentation, RecAug, RecConAug, RecResizeImg,

     RFLRecResizeImg, SVTRRecAug

 from .ssl_img_aug import SSLRotateResize

 from .randaugment import RandAugment

-from .copy_paste import CopyPaste

 from .ColorJitter import ColorJitter

 from .operators import *

 from .label_ops import *

  paddle2onnx = callPackage ../development/python-modules/paddle2onnx { };

  paddleocr = callPackage ../development/python-modules/paddleocr { };

  paddlepaddle = callPackage ../development/python-modules/paddlepaddle { };

  pulumi = callPackage ../development/python-modules/pulumi { inherit (pkgs) pulumi; };