python-doctr 0.8.1__tar.gz → 0.10.0__tar.gz

{python-doctr-0.8.1/python_doctr.egg-info → python_doctr-0.10.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: python-doctr
-Version: 0.8.1
+Version: 0.10.0
 Summary: Document Text Recognition (docTR): deep Learning for high-performance OCR on documents.
 Author-email: Mindee <contact@mindee.com>
 Maintainer: François-Guillaume Fernandez, Charles Gaillard, Olivier Dulcy, Felix Dittrich
@@ -219,42 +219,45 @@ Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Natural Language :: English
 Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: <4,>=3.8.0
+Requires-Python: <4,>=3.9.0
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: importlib_metadata
-Requires-Dist: numpy<2.0.0,>=1.16.0
+Requires-Dist: numpy<3.0.0,>=1.16.0
 Requires-Dist: scipy<2.0.0,>=1.4.0
 Requires-Dist: h5py<4.0.0,>=3.1.0
 Requires-Dist: opencv-python<5.0.0,>=4.5.0
-Requires-Dist: pypdfium2<5.0.0,>=4.0.0
+Requires-Dist: pypdfium2<5.0.0,>=4.11.0
 Requires-Dist: pyclipper<2.0.0,>=1.2.0
 Requires-Dist: shapely<3.0.0,>=1.6.0
 Requires-Dist: langdetect<2.0.0,>=1.0.9
 Requires-Dist: rapidfuzz<4.0.0,>=3.0.0
 Requires-Dist: huggingface-hub<1.0.0,>=0.20.0
-Requires-Dist: matplotlib>=3.1.0
-Requires-Dist: weasyprint>=55.0
 Requires-Dist: Pillow>=9.2.0
 Requires-Dist: defusedxml>=0.7.0
-Requires-Dist: mplcursors>=0.3
-Requires-Dist: unidecode>=1.0.0
+Requires-Dist: anyascii>=0.3.2
 Requires-Dist: tqdm>=4.30.0
 Provides-Extra: tf
-Requires-Dist: tensorflow<2.16.0,>=2.11.0; extra == "tf"
+Requires-Dist: tensorflow<3.0.0,>=2.15.0; extra == "tf"
+Requires-Dist: tf-keras<3.0.0,>=2.15.0; extra == "tf"
 Requires-Dist: tf2onnx<2.0.0,>=1.16.0; extra == "tf"
 Provides-Extra: torch
-Requires-Dist: torch<3.0.0,>=1.12.0; extra == "torch"
-Requires-Dist: torchvision>=0.13.0; extra == "torch"
+Requires-Dist: torch<3.0.0,>=2.0.0; extra == "torch"
+Requires-Dist: torchvision>=0.15.0; extra == "torch"
 Requires-Dist: onnx<3.0.0,>=1.12.0; extra == "torch"
+Provides-Extra: html
+Requires-Dist: weasyprint>=55.0; extra == "html"
+Provides-Extra: viz
+Requires-Dist: matplotlib>=3.1.0; extra == "viz"
+Requires-Dist: mplcursors>=0.3; extra == "viz"
+Provides-Extra: contrib
+Requires-Dist: onnxruntime>=1.11.0; extra == "contrib"
 Provides-Extra: testing
 Requires-Dist: pytest>=5.3.2; extra == "testing"
 Requires-Dist: coverage[toml]>=4.5.4; extra == "testing"
-Requires-Dist: hdf5storage>=0.1.18; extra == "testing"
 Requires-Dist: onnxruntime>=1.11.0; extra == "testing"
 Requires-Dist: requests>=2.20.0; extra == "testing"
 Requires-Dist: psutil>=5.9.5; extra == "testing"
@@ -266,30 +269,33 @@ Provides-Extra: docs
 Requires-Dist: sphinx!=3.5.0,>=3.0.0; extra == "docs"
 Requires-Dist: sphinxemoji>=0.1.8; extra == "docs"
 Requires-Dist: sphinx-copybutton>=0.3.1; extra == "docs"
-Requires-Dist: docutils<0.21; extra == "docs"
+Requires-Dist: docutils<0.22; extra == "docs"
 Requires-Dist: recommonmark>=0.7.1; extra == "docs"
 Requires-Dist: sphinx-markdown-tables>=0.0.15; extra == "docs"
 Requires-Dist: sphinx-tabs>=3.3.0; extra == "docs"
 Requires-Dist: furo>=2022.3.4; extra == "docs"
 Provides-Extra: dev
-Requires-Dist: tensorflow<2.16.0,>=2.11.0; extra == "dev"
+Requires-Dist: tensorflow<3.0.0,>=2.15.0; extra == "dev"
+Requires-Dist: tf-keras<3.0.0,>=2.15.0; extra == "dev"
 Requires-Dist: tf2onnx<2.0.0,>=1.16.0; extra == "dev"
-Requires-Dist: torch<3.0.0,>=1.12.0; extra == "dev"
-Requires-Dist: torchvision>=0.13.0; extra == "dev"
+Requires-Dist: torch<3.0.0,>=2.0.0; extra == "dev"
+Requires-Dist: torchvision>=0.15.0; extra == "dev"
 Requires-Dist: onnx<3.0.0,>=1.12.0; extra == "dev"
+Requires-Dist: weasyprint>=55.0; extra == "dev"
+Requires-Dist: matplotlib>=3.1.0; extra == "dev"
+Requires-Dist: mplcursors>=0.3; extra == "dev"
 Requires-Dist: pytest>=5.3.2; extra == "dev"
 Requires-Dist: coverage[toml]>=4.5.4; extra == "dev"
-Requires-Dist: hdf5storage>=0.1.18; extra == "dev"
 Requires-Dist: onnxruntime>=1.11.0; extra == "dev"
 Requires-Dist: requests>=2.20.0; extra == "dev"
 Requires-Dist: psutil>=5.9.5; extra == "dev"
-Requires-Dist: ruff>=0.1.5; extra == "dev"
-Requires-Dist: mypy>=0.812; extra == "dev"
-Requires-Dist: pre-commit>=2.17.0; extra == "dev"
+Requires-Dist: ruff>=0.3.0; extra == "dev"
+Requires-Dist: mypy>=1.0; extra == "dev"
+Requires-Dist: pre-commit>=3.0.0; extra == "dev"
 Requires-Dist: sphinx!=3.5.0,>=3.0.0; extra == "dev"
 Requires-Dist: sphinxemoji>=0.1.8; extra == "dev"
 Requires-Dist: sphinx-copybutton>=0.3.1; extra == "dev"
-Requires-Dist: docutils<0.21; extra == "dev"
+Requires-Dist: docutils<0.22; extra == "dev"
 Requires-Dist: recommonmark>=0.7.1; extra == "dev"
 Requires-Dist: sphinx-markdown-tables>=0.0.15; extra == "dev"
 Requires-Dist: sphinx-tabs>=3.3.0; extra == "dev"
@@ -299,7 +305,7 @@ Requires-Dist: furo>=2022.3.4; extra == "dev"
   <img src="https://github.com/mindee/doctr/raw/main/docs/images/Logo_doctr.gif" width="40%">
 </p>
 
-[![Slack Icon](https://img.shields.io/badge/Slack-Community-4A154B?style=flat-square&logo=slack&logoColor=white)](https://slack.mindee.com) [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE) ![Build Status](https://github.com/mindee/doctr/workflows/builds/badge.svg) [![Docker Images](https://img.shields.io/badge/Docker-4287f5?style=flat&logo=docker&logoColor=white)](https://github.com/mindee/doctr/pkgs/container/doctr) [![codecov](https://codecov.io/gh/mindee/doctr/branch/main/graph/badge.svg?token=577MO567NM)](https://codecov.io/gh/mindee/doctr) [![CodeFactor](https://www.codefactor.io/repository/github/mindee/doctr/badge?s=bae07db86bb079ce9d6542315b8c6e70fa708a7e)](https://www.codefactor.io/repository/github/mindee/doctr) [![Codacy Badge](https://api.codacy.com/project/badge/Grade/340a76749b634586a498e1c0ab998f08)](https://app.codacy.com/gh/mindee/doctr?utm_source=github.com&utm_medium=referral&utm_content=mindee/doctr&utm_campaign=Badge_Grade) [![Doc Status](https://github.com/mindee/doctr/workflows/doc-status/badge.svg)](https://mindee.github.io/doctr) [![Pypi](https://img.shields.io/badge/pypi-v0.8.0-blue.svg)](https://pypi.org/project/python-doctr/) [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/mindee/doctr) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/mindee/notebooks/blob/main/doctr/quicktour.ipynb)
+[![Slack Icon](https://img.shields.io/badge/Slack-Community-4A154B?style=flat-square&logo=slack&logoColor=white)](https://slack.mindee.com) [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE) ![Build Status](https://github.com/mindee/doctr/workflows/builds/badge.svg) [![Docker Images](https://img.shields.io/badge/Docker-4287f5?style=flat&logo=docker&logoColor=white)](https://github.com/mindee/doctr/pkgs/container/doctr) [![codecov](https://codecov.io/gh/mindee/doctr/branch/main/graph/badge.svg?token=577MO567NM)](https://codecov.io/gh/mindee/doctr) [![CodeFactor](https://www.codefactor.io/repository/github/mindee/doctr/badge?s=bae07db86bb079ce9d6542315b8c6e70fa708a7e)](https://www.codefactor.io/repository/github/mindee/doctr) [![Codacy Badge](https://api.codacy.com/project/badge/Grade/340a76749b634586a498e1c0ab998f08)](https://app.codacy.com/gh/mindee/doctr?utm_source=github.com&utm_medium=referral&utm_content=mindee/doctr&utm_campaign=Badge_Grade) [![Doc Status](https://github.com/mindee/doctr/workflows/doc-status/badge.svg)](https://mindee.github.io/doctr) [![Pypi](https://img.shields.io/badge/pypi-v0.9.0-blue.svg)](https://pypi.org/project/python-doctr/) [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/mindee/doctr) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/mindee/notebooks/blob/main/doctr/quicktour.ipynb)
 
 
 **Optical Character Recognition made seamless & accessible to anyone, powered by TensorFlow 2 & PyTorch**
@@ -334,7 +340,7 @@ from doctr.io import DocumentFile
 pdf_doc = DocumentFile.from_pdf("path/to/your/doc.pdf")
 # Image
 single_img_doc = DocumentFile.from_images("path/to/your/img.jpg")
-# Webpage
+# Webpage (requires `weasyprint` to be installed)
 webpage_doc = DocumentFile.from_url("https://www.yoursite.com")
 # Multiple page images
 multi_img_doc = DocumentFile.from_images(["path/to/page1.jpg", "path/to/page2.jpg"])
@@ -372,6 +378,7 @@ If both options are set to False, the predictor will always fit and return rotat
 To interpret your model's predictions, you can visualize them interactively as follows:
 
 ```python
+# Display the result (requires matplotlib & mplcursors to be installed)
 result.show()
 ```
 
@@ -431,17 +438,7 @@ The KIE predictor results per page are in a dictionary format with each key repr
 
 ### Prerequisites
 
-Python 3.8 (or higher) and [pip](https://pip.pypa.io/en/stable/) are required to install docTR.
-
-Since we use [weasyprint](https://weasyprint.org/), you will need extra dependencies if you are not running Linux.
-
-For MacOS users, you can install them as follows:
-
-```shell
-brew install cairo pango gdk-pixbuf libffi
-```
-
-For Windows users, those dependencies are included in GTK. You can find the latest installer over [here](https://github.com/tschoonj/GTK-for-Windows-Runtime-Environment-Installer/releases).
+Python 3.9 (or higher) and [pip](https://pip.pypa.io/en/stable/) are required to install docTR.
 
 ### Latest release
 
@@ -460,12 +457,14 @@ We try to keep framework-specific dependencies to a minimum. You can install fra
 pip install "python-doctr[tf]"
 # for PyTorch
 pip install "python-doctr[torch]"
+# optional dependencies for visualization, html, and contrib modules can be installed as follows:
+pip install "python-doctr[torch,viz,html,contib]"
 ```
 
 For MacBooks with M1 chip, you will need some additional packages or specific versions:
 
 - TensorFlow 2: [metal plugin](https://developer.apple.com/metal/tensorflow-plugin/)
-- PyTorch: [version >= 1.12.0](https://pytorch.org/get-started/locally/#start-locally)
+- PyTorch: [version >= 2.0.0](https://pytorch.org/get-started/locally/#start-locally)
 
 ### Developer mode
 
@@ -647,9 +646,14 @@ Your API should now be running locally on your port 8002. Access your automatica
 
 ```python
 import requests
+
+params = {"det_arch": "db_resnet50", "reco_arch": "crnn_vgg16_bn"}
+
 with open('/path/to/your/doc.jpg', 'rb') as f:
-    data = f.read()
-response = requests.post("http://localhost:8002/ocr", files={'file': data}).json()
+    files = [  # application/pdf, image/jpeg, image/png supported
+        ("files", ("doc.jpg", f.read(), "image/jpeg")),
+    ]
+print(requests.post("http://localhost:8080/ocr", params=params, files=files).json())
 ```
 
 ### Example notebooks

{python-doctr-0.8.1 → python_doctr-0.10.0}/README.md

@@ -2,7 +2,7 @@
   <img src="https://github.com/mindee/doctr/raw/main/docs/images/Logo_doctr.gif" width="40%">
 </p>
 
-[![Slack Icon](https://img.shields.io/badge/Slack-Community-4A154B?style=flat-square&logo=slack&logoColor=white)](https://slack.mindee.com) [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE) ![Build Status](https://github.com/mindee/doctr/workflows/builds/badge.svg) [![Docker Images](https://img.shields.io/badge/Docker-4287f5?style=flat&logo=docker&logoColor=white)](https://github.com/mindee/doctr/pkgs/container/doctr) [![codecov](https://codecov.io/gh/mindee/doctr/branch/main/graph/badge.svg?token=577MO567NM)](https://codecov.io/gh/mindee/doctr) [![CodeFactor](https://www.codefactor.io/repository/github/mindee/doctr/badge?s=bae07db86bb079ce9d6542315b8c6e70fa708a7e)](https://www.codefactor.io/repository/github/mindee/doctr) [![Codacy Badge](https://api.codacy.com/project/badge/Grade/340a76749b634586a498e1c0ab998f08)](https://app.codacy.com/gh/mindee/doctr?utm_source=github.com&utm_medium=referral&utm_content=mindee/doctr&utm_campaign=Badge_Grade) [![Doc Status](https://github.com/mindee/doctr/workflows/doc-status/badge.svg)](https://mindee.github.io/doctr) [![Pypi](https://img.shields.io/badge/pypi-v0.8.0-blue.svg)](https://pypi.org/project/python-doctr/) [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/mindee/doctr) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/mindee/notebooks/blob/main/doctr/quicktour.ipynb)
+[![Slack Icon](https://img.shields.io/badge/Slack-Community-4A154B?style=flat-square&logo=slack&logoColor=white)](https://slack.mindee.com) [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE) ![Build Status](https://github.com/mindee/doctr/workflows/builds/badge.svg) [![Docker Images](https://img.shields.io/badge/Docker-4287f5?style=flat&logo=docker&logoColor=white)](https://github.com/mindee/doctr/pkgs/container/doctr) [![codecov](https://codecov.io/gh/mindee/doctr/branch/main/graph/badge.svg?token=577MO567NM)](https://codecov.io/gh/mindee/doctr) [![CodeFactor](https://www.codefactor.io/repository/github/mindee/doctr/badge?s=bae07db86bb079ce9d6542315b8c6e70fa708a7e)](https://www.codefactor.io/repository/github/mindee/doctr) [![Codacy Badge](https://api.codacy.com/project/badge/Grade/340a76749b634586a498e1c0ab998f08)](https://app.codacy.com/gh/mindee/doctr?utm_source=github.com&utm_medium=referral&utm_content=mindee/doctr&utm_campaign=Badge_Grade) [![Doc Status](https://github.com/mindee/doctr/workflows/doc-status/badge.svg)](https://mindee.github.io/doctr) [![Pypi](https://img.shields.io/badge/pypi-v0.9.0-blue.svg)](https://pypi.org/project/python-doctr/) [![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/mindee/doctr) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/mindee/notebooks/blob/main/doctr/quicktour.ipynb)
 
 
 **Optical Character Recognition made seamless & accessible to anyone, powered by TensorFlow 2 & PyTorch**
@@ -37,7 +37,7 @@ from doctr.io import DocumentFile
 pdf_doc = DocumentFile.from_pdf("path/to/your/doc.pdf")
 # Image
 single_img_doc = DocumentFile.from_images("path/to/your/img.jpg")
-# Webpage
+# Webpage (requires `weasyprint` to be installed)
 webpage_doc = DocumentFile.from_url("https://www.yoursite.com")
 # Multiple page images
 multi_img_doc = DocumentFile.from_images(["path/to/page1.jpg", "path/to/page2.jpg"])
@@ -75,6 +75,7 @@ If both options are set to False, the predictor will always fit and return rotat
 To interpret your model's predictions, you can visualize them interactively as follows:
 
 ```python
+# Display the result (requires matplotlib & mplcursors to be installed)
 result.show()
 ```
 
@@ -134,17 +135,7 @@ The KIE predictor results per page are in a dictionary format with each key repr
 
 ### Prerequisites
 
-Python 3.8 (or higher) and [pip](https://pip.pypa.io/en/stable/) are required to install docTR.
-
-Since we use [weasyprint](https://weasyprint.org/), you will need extra dependencies if you are not running Linux.
-
-For MacOS users, you can install them as follows:
-
-```shell
-brew install cairo pango gdk-pixbuf libffi
-```
-
-For Windows users, those dependencies are included in GTK. You can find the latest installer over [here](https://github.com/tschoonj/GTK-for-Windows-Runtime-Environment-Installer/releases).
+Python 3.9 (or higher) and [pip](https://pip.pypa.io/en/stable/) are required to install docTR.
 
 ### Latest release
 
@@ -163,12 +154,14 @@ We try to keep framework-specific dependencies to a minimum. You can install fra
 pip install "python-doctr[tf]"
 # for PyTorch
 pip install "python-doctr[torch]"
+# optional dependencies for visualization, html, and contrib modules can be installed as follows:
+pip install "python-doctr[torch,viz,html,contib]"
 ```
 
 For MacBooks with M1 chip, you will need some additional packages or specific versions:
 
 - TensorFlow 2: [metal plugin](https://developer.apple.com/metal/tensorflow-plugin/)
-- PyTorch: [version >= 1.12.0](https://pytorch.org/get-started/locally/#start-locally)
+- PyTorch: [version >= 2.0.0](https://pytorch.org/get-started/locally/#start-locally)
 
 ### Developer mode
 
@@ -350,9 +343,14 @@ Your API should now be running locally on your port 8002. Access your automatica
 
 ```python
 import requests
+
+params = {"det_arch": "db_resnet50", "reco_arch": "crnn_vgg16_bn"}
+
 with open('/path/to/your/doc.jpg', 'rb') as f:
-    data = f.read()
-response = requests.post("http://localhost:8002/ocr", files={'file': data}).json()
+    files = [  # application/pdf, image/jpeg, image/png supported
+        ("files", ("doc.jpg", f.read(), "image/jpeg")),
+    ]
+print(requests.post("http://localhost:8080/ocr", params=params, files=files).json())
 ```
 
 ### Example notebooks

{python-doctr-0.8.1 → python_doctr-0.10.0}/doctr/__init__.py

@@ -1,3 +1,3 @@
-from . import io, models, datasets, transforms, utils
+from . import io, models, datasets, contrib, transforms, utils
 from .file_utils import is_tf_available, is_torch_available
 from .version import __version__  # noqa: F401

python_doctr-0.10.0/doctr/contrib/artefacts.py

@@ -0,0 +1,131 @@
+# Copyright (C) 2021-2024, Mindee.
+
+# This program is licensed under the Apache License 2.0.
+# See LICENSE or go to <https://opensource.org/licenses/Apache-2.0> for full license details.
+
+from typing import Any, Dict, List, Optional, Tuple
+
+import cv2
+import numpy as np
+
+from doctr.file_utils import requires_package
+
+from .base import _BasePredictor
+
+__all__ = ["ArtefactDetector"]
+
+default_cfgs: Dict[str, Dict[str, Any]] = {
+    "yolov8_artefact": {
+        "input_shape": (3, 1024, 1024),
+        "labels": ["bar_code", "qr_code", "logo", "photo"],
+        "url": "https://doctr-static.mindee.com/models?id=v0.8.1/yolo_artefact-f9d66f14.onnx&src=0",
+    },
+}
+
+
+class ArtefactDetector(_BasePredictor):
+    """
+    A class to detect artefacts in images
+
+    >>> from doctr.io import DocumentFile
+    >>> from doctr.contrib.artefacts import ArtefactDetector
+    >>> doc = DocumentFile.from_images(["path/to/image.jpg"])
+    >>> detector = ArtefactDetector()
+    >>> results = detector(doc)
+
+    Args:
+    ----
+        arch: the architecture to use
+        batch_size: the batch size to use
+        model_path: the path to the model to use
+        labels: the labels to use
+        input_shape: the input shape to use
+        mask_labels: the mask labels to use
+        conf_threshold: the confidence threshold to use
+        iou_threshold: the intersection over union threshold to use
+        **kwargs: additional arguments to be passed to `download_from_url`
+    """
+
+    def __init__(
+        self,
+        arch: str = "yolov8_artefact",
+        batch_size: int = 2,
+        model_path: Optional[str] = None,
+        labels: Optional[List[str]] = None,
+        input_shape: Optional[Tuple[int, int, int]] = None,
+        conf_threshold: float = 0.5,
+        iou_threshold: float = 0.5,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__(batch_size=batch_size, url=default_cfgs[arch]["url"], model_path=model_path, **kwargs)
+        self.labels = labels or default_cfgs[arch]["labels"]
+        self.input_shape = input_shape or default_cfgs[arch]["input_shape"]
+        self.conf_threshold = conf_threshold
+        self.iou_threshold = iou_threshold
+
+    def preprocess(self, img: np.ndarray) -> np.ndarray:
+        return np.transpose(cv2.resize(img, (self.input_shape[2], self.input_shape[1])), (2, 0, 1)) / np.array(255.0)
+
+    def postprocess(self, output: List[np.ndarray], input_images: List[List[np.ndarray]]) -> List[List[Dict[str, Any]]]:
+        results = []
+
+        for batch in zip(output, input_images):
+            for out, img in zip(batch[0], batch[1]):
+                org_height, org_width = img.shape[:2]
+                width_scale, height_scale = org_width / self.input_shape[2], org_height / self.input_shape[1]
+                for res in out:
+                    sample_results = []
+                    for row in np.transpose(np.squeeze(res)):
+                        classes_scores = row[4:]
+                        max_score = np.amax(classes_scores)
+                        if max_score >= self.conf_threshold:
+                            class_id = np.argmax(classes_scores)
+                            x, y, w, h = row[0], row[1], row[2], row[3]
+                            # to rescaled xmin, ymin, xmax, ymax
+                            xmin = int((x - w / 2) * width_scale)
+                            ymin = int((y - h / 2) * height_scale)
+                            xmax = int((x + w / 2) * width_scale)
+                            ymax = int((y + h / 2) * height_scale)
+
+                            sample_results.append({
+                                "label": self.labels[class_id],
+                                "confidence": float(max_score),
+                                "box": [xmin, ymin, xmax, ymax],
+                            })
+
+                    # Filter out overlapping boxes
+                    boxes = [res["box"] for res in sample_results]
+                    scores = [res["confidence"] for res in sample_results]
+                    keep_indices = cv2.dnn.NMSBoxes(boxes, scores, self.conf_threshold, self.iou_threshold)  # type: ignore[arg-type]
+                    sample_results = [sample_results[i] for i in keep_indices]
+
+                    results.append(sample_results)
+
+        self._results = results
+        return results
+
+    def show(self, **kwargs: Any) -> None:
+        """
+        Display the results
+
+        Args:
+        ----
+            **kwargs: additional keyword arguments to be passed to `plt.show`
+        """
+        requires_package("matplotlib", "`.show()` requires matplotlib installed")
+        import matplotlib.pyplot as plt
+        from matplotlib.patches import Rectangle
+
+        # visualize the results with matplotlib
+        if self._results and self._inputs:
+            for img, res in zip(self._inputs, self._results):
+                plt.figure(figsize=(10, 10))
+                plt.imshow(img)
+                for obj in res:
+                    xmin, ymin, xmax, ymax = obj["box"]
+                    label = obj["label"]
+                    plt.text(xmin, ymin, f"{label} {obj['confidence']:.2f}", color="red")
+                    plt.gca().add_patch(
+                        Rectangle((xmin, ymin), xmax - xmin, ymax - ymin, fill=False, edgecolor="red", linewidth=2)
+                    )
+                plt.show(**kwargs)

python_doctr-0.10.0/doctr/contrib/base.py

@@ -0,0 +1,105 @@
+# Copyright (C) 2021-2024, Mindee.
+
+# This program is licensed under the Apache License 2.0.
+# See LICENSE or go to <https://opensource.org/licenses/Apache-2.0> for full license details.
+
+from typing import Any, List, Optional
+
+import numpy as np
+
+from doctr.file_utils import requires_package
+from doctr.utils.data import download_from_url
+
+
+class _BasePredictor:
+    """
+    Base class for all predictors
+
+    Args:
+    ----
+        batch_size: the batch size to use
+        url: the url to use to download a model if needed
+        model_path: the path to the model to use
+        **kwargs: additional arguments to be passed to `download_from_url`
+    """
+
+    def __init__(self, batch_size: int, url: Optional[str] = None, model_path: Optional[str] = None, **kwargs) -> None:
+        self.batch_size = batch_size
+        self.session = self._init_model(url, model_path, **kwargs)
+
+        self._inputs: List[np.ndarray] = []
+        self._results: List[Any] = []
+
+    def _init_model(self, url: Optional[str] = None, model_path: Optional[str] = None, **kwargs: Any) -> Any:
+        """
+        Download the model from the given url if needed
+
+        Args:
+        ----
+            url: the url to use
+            model_path: the path to the model to use
+            **kwargs: additional arguments to be passed to `download_from_url`
+
+        Returns:
+        -------
+            Any: the ONNX loaded model
+        """
+        requires_package("onnxruntime", "`.contrib` module requires `onnxruntime` to be installed.")
+        import onnxruntime as ort
+
+        if not url and not model_path:
+            raise ValueError("You must provide either a url or a model_path")
+        onnx_model_path = model_path if model_path else str(download_from_url(url, cache_subdir="models", **kwargs))  # type: ignore[arg-type]
+        return ort.InferenceSession(onnx_model_path, providers=["CUDAExecutionProvider", "CPUExecutionProvider"])
+
+    def preprocess(self, img: np.ndarray) -> np.ndarray:
+        """
+        Preprocess the input image
+
+        Args:
+        ----
+            img: the input image to preprocess
+
+        Returns:
+        -------
+            np.ndarray: the preprocessed image
+        """
+        raise NotImplementedError
+
+    def postprocess(self, output: List[np.ndarray], input_images: List[List[np.ndarray]]) -> Any:
+        """
+        Postprocess the model output
+
+        Args:
+        ----
+            output: the model output to postprocess
+            input_images: the input images used to generate the output
+
+        Returns:
+        -------
+            Any: the postprocessed output
+        """
+        raise NotImplementedError
+
+    def __call__(self, inputs: List[np.ndarray]) -> Any:
+        """
+        Call the model on the given inputs
+
+        Args:
+        ----
+            inputs: the inputs to use
+
+        Returns:
+        -------
+            Any: the postprocessed output
+        """
+        self._inputs = inputs
+        model_inputs = self.session.get_inputs()
+
+        batched_inputs = [inputs[i : i + self.batch_size] for i in range(0, len(inputs), self.batch_size)]
+        processed_batches = [
+            np.array([self.preprocess(img) for img in batch], dtype=np.float32) for batch in batched_inputs
+        ]
+
+        outputs = [self.session.run(None, {model_inputs[0].name: batch}) for batch in processed_batches]
+        return self.postprocess(outputs, batched_inputs)

{python-doctr-0.8.1 → python_doctr-0.10.0}/doctr/datasets/cord.py

@@ -33,6 +33,7 @@ class CORD(VisionDataset):
         train: whether the subset should be the training one
         use_polygons: whether polygons should be considered as rotated bounding box (instead of straight ones)
         recognition_task: whether the dataset should be used for recognition task
+        detection_task: whether the dataset should be used for detection task
         **kwargs: keyword arguments from `VisionDataset`.
     """
 
@@ -53,6 +54,7 @@ class CORD(VisionDataset):
         train: bool = True,
         use_polygons: bool = False,
         recognition_task: bool = False,
+        detection_task: bool = False,
         **kwargs: Any,
     ) -> None:
         url, sha256, name = self.TRAIN if train else self.TEST
@@ -64,10 +66,15 @@ class CORD(VisionDataset):
             pre_transforms=convert_target_to_relative if not recognition_task else None,
             **kwargs,
         )
+        if recognition_task and detection_task:
+            raise ValueError(
+                "`recognition_task` and `detection_task` cannot be set to True simultaneously. "
+                + "To get the whole dataset with boxes and labels leave both parameters to False."
+            )
 
         # List images
         tmp_root = os.path.join(self.root, "image")
-        self.data: List[Tuple[Union[str, np.ndarray], Union[str, Dict[str, Any]]]] = []
+        self.data: List[Tuple[Union[str, np.ndarray], Union[str, Dict[str, Any], np.ndarray]]] = []
         self.train = train
         np_dtype = np.float32
         for img_path in tqdm(iterable=os.listdir(tmp_root), desc="Unpacking CORD", total=len(os.listdir(tmp_root))):
@@ -109,6 +116,8 @@ class CORD(VisionDataset):
                 )
                 for crop, label in zip(crops, list(text_targets)):
                     self.data.append((crop, label))
+            elif detection_task:
+                self.data.append((img_path, np.asarray(box_targets, dtype=int).clip(min=0)))
             else:
                 self.data.append((
                     img_path,

{python-doctr-0.8.1 → python_doctr-0.10.0}/doctr/datasets/datasets/pytorch.py

@@ -50,9 +50,9 @@ class AbstractDataset(_AbstractDataset):
     @staticmethod
     def collate_fn(samples: List[Tuple[torch.Tensor, Any]]) -> Tuple[torch.Tensor, List[Any]]:
         images, targets = zip(*samples)
-        images = torch.stack(images, dim=0)
+        images = torch.stack(images, dim=0)  # type: ignore[assignment]
 
-        return images, list(targets)
+        return images, list(targets)  # type: ignore[return-value]
 
 
 class VisionDataset(AbstractDataset, _VisionDataset):  # noqa: D101

{python-doctr-0.8.1 → python_doctr-0.10.0}/doctr/datasets/funsd.py

@@ -33,6 +33,7 @@ class FUNSD(VisionDataset):
         train: whether the subset should be the training one
         use_polygons: whether polygons should be considered as rotated bounding box (instead of straight ones)
         recognition_task: whether the dataset should be used for recognition task
+        detection_task: whether the dataset should be used for detection task
         **kwargs: keyword arguments from `VisionDataset`.
     """
 
@@ -45,6 +46,7 @@ class FUNSD(VisionDataset):
         train: bool = True,
         use_polygons: bool = False,
         recognition_task: bool = False,
+        detection_task: bool = False,
         **kwargs: Any,
     ) -> None:
         super().__init__(
@@ -55,6 +57,12 @@ class FUNSD(VisionDataset):
             pre_transforms=convert_target_to_relative if not recognition_task else None,
             **kwargs,
         )
+        if recognition_task and detection_task:
+            raise ValueError(
+                "`recognition_task` and `detection_task` cannot be set to True simultaneously. "
+                + "To get the whole dataset with boxes and labels leave both parameters to False."
+            )
+
         self.train = train
         np_dtype = np.float32
 
@@ -63,7 +71,7 @@ class FUNSD(VisionDataset):
 
         # # List images
         tmp_root = os.path.join(self.root, subfolder, "images")
-        self.data: List[Tuple[Union[str, np.ndarray], Union[str, Dict[str, Any]]]] = []
+        self.data: List[Tuple[Union[str, np.ndarray], Union[str, Dict[str, Any], np.ndarray]]] = []
         for img_path in tqdm(iterable=os.listdir(tmp_root), desc="Unpacking FUNSD", total=len(os.listdir(tmp_root))):
             # File existence check
             if not os.path.exists(os.path.join(tmp_root, img_path)):
@@ -100,6 +108,8 @@ class FUNSD(VisionDataset):
                     # filter labels with unknown characters
                     if not any(char in label for char in ["☑", "☐", "\uf703", "\uf702"]):
                         self.data.append((crop, label))
+            elif detection_task:
+                self.data.append((img_path, np.asarray(box_targets, dtype=np_dtype)))
             else:
                 self.data.append((
                     img_path,