python-doctr 0.8.1__py3-none-any.whl → 0.10.0__py3-none-any.whl

doctr/utils/reconstitution.py

@@ -0,0 +1,212 @@
+# 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.
+import logging
+from typing import Any, Dict, Optional
+
+import numpy as np
+from anyascii import anyascii
+from PIL import Image, ImageDraw
+
+from .fonts import get_font
+
+__all__ = ["synthesize_page", "synthesize_kie_page"]
+
+
+# Global variable to avoid multiple warnings
+ROTATION_WARNING = False
+
+
+def _warn_rotation(entry: Dict[str, Any]) -> None:  # pragma: no cover
+    global ROTATION_WARNING
+    if not ROTATION_WARNING and len(entry["geometry"]) == 4:
+        logging.warning("Polygons with larger rotations will lead to inaccurate rendering")
+        ROTATION_WARNING = True
+
+
+def _synthesize(
+    response: Image.Image,
+    entry: Dict[str, Any],
+    w: int,
+    h: int,
+    draw_proba: bool = False,
+    font_family: Optional[str] = None,
+    smoothing_factor: float = 0.75,
+    min_font_size: int = 6,
+    max_font_size: int = 50,
+) -> Image.Image:
+    if len(entry["geometry"]) == 2:
+        (xmin, ymin), (xmax, ymax) = entry["geometry"]
+        polygon = [(xmin, ymin), (xmax, ymin), (xmax, ymax), (xmin, ymax)]
+    else:
+        polygon = entry["geometry"]
+
+    # Calculate the bounding box of the word
+    x_coords, y_coords = zip(*polygon)
+    xmin, ymin, xmax, ymax = (
+        int(round(w * min(x_coords))),
+        int(round(h * min(y_coords))),
+        int(round(w * max(x_coords))),
+        int(round(h * max(y_coords))),
+    )
+    word_width = xmax - xmin
+    word_height = ymax - ymin
+
+    # If lines are provided instead of words, concatenate the word entries
+    if "words" in entry:
+        word_text = " ".join(word["value"] for word in entry["words"])
+    else:
+        word_text = entry["value"]
+    # Find the optimal font size
+    try:
+        font_size = min(word_height, max_font_size)
+        font = get_font(font_family, font_size)
+        text_width, text_height = font.getbbox(word_text)[2:4]
+
+        while (text_width > word_width or text_height > word_height) and font_size > min_font_size:
+            font_size = max(int(font_size * smoothing_factor), min_font_size)
+            font = get_font(font_family, font_size)
+            text_width, text_height = font.getbbox(word_text)[2:4]
+    except ValueError:
+        font = get_font(font_family, min_font_size)
+
+    # Create a mask for the word
+    mask = Image.new("L", (w, h), 0)
+    ImageDraw.Draw(mask).polygon([(int(round(w * x)), int(round(h * y))) for x, y in polygon], fill=255)
+
+    # Draw the word text
+    d = ImageDraw.Draw(response)
+    try:
+        try:
+            d.text((xmin, ymin), word_text, font=font, fill=(0, 0, 0), anchor="lt")
+        except UnicodeEncodeError:
+            d.text((xmin, ymin), anyascii(word_text), font=font, fill=(0, 0, 0), anchor="lt")
+    # Catch generic exceptions to avoid crashing the whole rendering
+    except Exception:  # pragma: no cover
+        logging.warning(f"Could not render word: {word_text}")
+
+    if draw_proba:
+        confidence = (
+            entry["confidence"]
+            if "confidence" in entry
+            else sum(w["confidence"] for w in entry["words"]) / len(entry["words"])
+        )
+        p = int(255 * confidence)
+        color = (255 - p, 0, p)  # Red to blue gradient based on probability
+        d.rectangle([(xmin, ymin), (xmax, ymax)], outline=color, width=2)
+
+        prob_font = get_font(font_family, 20)
+        prob_text = f"{confidence:.2f}"
+        prob_text_width, prob_text_height = prob_font.getbbox(prob_text)[2:4]
+
+        # Position the probability slightly above the bounding box
+        prob_x_offset = (word_width - prob_text_width) // 2
+        prob_y_offset = ymin - prob_text_height - 2
+        prob_y_offset = max(0, prob_y_offset)
+
+        d.text((xmin + prob_x_offset, prob_y_offset), prob_text, font=prob_font, fill=color, anchor="lt")
+
+    return response
+
+
+def synthesize_page(
+    page: Dict[str, Any],
+    draw_proba: bool = False,
+    font_family: Optional[str] = None,
+    smoothing_factor: float = 0.95,
+    min_font_size: int = 8,
+    max_font_size: int = 50,
+) -> np.ndarray:
+    """Draw a the content of the element page (OCR response) on a blank page.
+
+    Args:
+    ----
+        page: exported Page object to represent
+        draw_proba: if True, draw words in colors to represent confidence. Blue: p=1, red: p=0
+        font_family: family of the font
+        smoothing_factor: factor to smooth the font size
+        min_font_size: minimum font size
+        max_font_size: maximum font size
+
+    Returns:
+    -------
+        the synthesized page
+    """
+    # Draw template
+    h, w = page["dimensions"]
+    response = Image.new("RGB", (w, h), color=(255, 255, 255))
+
+    for block in page["blocks"]:
+        # If lines are provided use these to get better rendering results
+        if len(block["lines"]) > 1:
+            for line in block["lines"]:
+                _warn_rotation(block)  # pragma: no cover
+                response = _synthesize(
+                    response=response,
+                    entry=line,
+                    w=w,
+                    h=h,
+                    draw_proba=draw_proba,
+                    font_family=font_family,
+                    smoothing_factor=smoothing_factor,
+                    min_font_size=min_font_size,
+                    max_font_size=max_font_size,
+                )
+        # Otherwise, draw each word
+        else:
+            for line in block["lines"]:
+                _warn_rotation(block)  # pragma: no cover
+                for word in line["words"]:
+                    response = _synthesize(
+                        response=response,
+                        entry=word,
+                        w=w,
+                        h=h,
+                        draw_proba=draw_proba,
+                        font_family=font_family,
+                        smoothing_factor=smoothing_factor,
+                        min_font_size=min_font_size,
+                        max_font_size=max_font_size,
+                    )
+
+    return np.array(response, dtype=np.uint8)
+
+
+def synthesize_kie_page(
+    page: Dict[str, Any],
+    draw_proba: bool = False,
+    font_family: Optional[str] = None,
+) -> np.ndarray:
+    """Draw a the content of the element page (OCR response) on a blank page.
+
+    Args:
+    ----
+        page: exported Page object to represent
+        draw_proba: if True, draw words in colors to represent confidence. Blue: p=1, red: p=0
+        font_family: family of the font
+        smoothing_factor: factor to smooth the font size
+        min_font_size: minimum font size
+        max_font_size: maximum font size
+
+    Returns:
+    -------
+        the synthesized page
+    """
+    # Draw template
+    h, w = page["dimensions"]
+    response = Image.new("RGB", (w, h), color=(255, 255, 255))
+
+    # Draw each word
+    for predictions in page["predictions"].values():
+        for prediction in predictions:
+            _warn_rotation(prediction)  # pragma: no cover
+            response = _synthesize(
+                response=response,
+                entry=prediction,
+                w=w,
+                h=h,
+                draw_proba=draw_proba,
+                font_family=font_family,
+            )
+    return np.array(response, dtype=np.uint8)

doctr/utils/visualization.py

@@ -9,16 +9,12 @@ from typing import Any, Dict, List, Optional, Tuple, Union
 import cv2
 import matplotlib.patches as patches
 import matplotlib.pyplot as plt
-import mplcursors
 import numpy as np
 from matplotlib.figure import Figure
-from PIL import Image, ImageDraw
-from unidecode import unidecode
 
 from .common_types import BoundingBox, Polygon4P
-from .fonts import get_font
 
-__all__ = ["visualize_page", "synthesize_page", "visualize_kie_page", "synthesize_kie_page", "draw_boxes"]
+__all__ = ["visualize_page", "visualize_kie_page", "draw_boxes"]
 
 
 def rect_patch(
@@ -281,6 +277,8 @@ def visualize_page(
                     artists.append(rect)
 
     if interactive:
+        import mplcursors
+
         # Create mlp Cursor to hover patches in artists
         mplcursors.Cursor(artists, hover=2).connect("add", lambda sel: sel.annotation.set_text(sel.artist.get_label()))
     fig.tight_layout(pad=0.0)
@@ -288,63 +286,6 @@ def visualize_page(
     return fig
 
 
-def synthesize_page(
-    page: Dict[str, Any],
-    draw_proba: bool = False,
-    font_family: Optional[str] = None,
-) -> np.ndarray:
-    """Draw a the content of the element page (OCR response) on a blank page.
-
-    Args:
-    ----
-        page: exported Page object to represent
-        draw_proba: if True, draw words in colors to represent confidence. Blue: p=1, red: p=0
-        font_size: size of the font, default font = 13
-        font_family: family of the font
-
-    Returns:
-    -------
-        the synthesized page
-    """
-    # Draw template
-    h, w = page["dimensions"]
-    response = 255 * np.ones((h, w, 3), dtype=np.int32)
-
-    # Draw each word
-    for block in page["blocks"]:
-        for line in block["lines"]:
-            for word in line["words"]:
-                # Get aboslute word geometry
-                (xmin, ymin), (xmax, ymax) = word["geometry"]
-                xmin, xmax = int(round(w * xmin)), int(round(w * xmax))
-                ymin, ymax = int(round(h * ymin)), int(round(h * ymax))
-
-                # White drawing context adapted to font size, 0.75 factor to convert pts --> pix
-                font = get_font(font_family, int(0.75 * (ymax - ymin)))
-                img = Image.new("RGB", (xmax - xmin, ymax - ymin), color=(255, 255, 255))
-                d = ImageDraw.Draw(img)
-                # Draw in black the value of the word
-                try:
-                    d.text((0, 0), word["value"], font=font, fill=(0, 0, 0))
-                except UnicodeEncodeError:
-                    # When character cannot be encoded, use its unidecode version
-                    d.text((0, 0), unidecode(word["value"]), font=font, fill=(0, 0, 0))
-
-                # Colorize if draw_proba
-                if draw_proba:
-                    p = int(255 * word["confidence"])
-                    mask = np.where(np.array(img) == 0, 1, 0)
-                    proba: np.ndarray = np.array([255 - p, 0, p])
-                    color = mask * proba[np.newaxis, np.newaxis, :]
-                    white_mask = 255 * (1 - mask)
-                    img = color + white_mask
-
-                # Write to response page
-                response[ymin:ymax, xmin:xmax, :] = np.array(img)
-
-    return response
-
-
 def visualize_kie_page(
     page: Dict[str, Any],
     image: np.ndarray,
@@ -413,6 +354,8 @@ def visualize_kie_page(
                     artists.append(rect)
 
     if interactive:
+        import mplcursors
+
         # Create mlp Cursor to hover patches in artists
         mplcursors.Cursor(artists, hover=2).connect("add", lambda sel: sel.annotation.set_text(sel.artist.get_label()))
     fig.tight_layout(pad=0.0)
@@ -420,62 +363,6 @@ def visualize_kie_page(
     return fig
 
 
-def synthesize_kie_page(
-    page: Dict[str, Any],
-    draw_proba: bool = False,
-    font_family: Optional[str] = None,
-) -> np.ndarray:
-    """Draw a the content of the element page (OCR response) on a blank page.
-
-    Args:
-    ----
-        page: exported Page object to represent
-        draw_proba: if True, draw words in colors to represent confidence. Blue: p=1, red: p=0
-        font_size: size of the font, default font = 13
-        font_family: family of the font
-
-    Returns:
-    -------
-        the synthesized page
-    """
-    # Draw template
-    h, w = page["dimensions"]
-    response = 255 * np.ones((h, w, 3), dtype=np.int32)
-
-    # Draw each word
-    for predictions in page["predictions"].values():
-        for prediction in predictions:
-            # Get aboslute word geometry
-            (xmin, ymin), (xmax, ymax) = prediction["geometry"]
-            xmin, xmax = int(round(w * xmin)), int(round(w * xmax))
-            ymin, ymax = int(round(h * ymin)), int(round(h * ymax))
-
-            # White drawing context adapted to font size, 0.75 factor to convert pts --> pix
-            font = get_font(font_family, int(0.75 * (ymax - ymin)))
-            img = Image.new("RGB", (xmax - xmin, ymax - ymin), color=(255, 255, 255))
-            d = ImageDraw.Draw(img)
-            # Draw in black the value of the word
-            try:
-                d.text((0, 0), prediction["value"], font=font, fill=(0, 0, 0))
-            except UnicodeEncodeError:
-                # When character cannot be encoded, use its unidecode version
-                d.text((0, 0), unidecode(prediction["value"]), font=font, fill=(0, 0, 0))
-
-            # Colorize if draw_proba
-            if draw_proba:
-                p = int(255 * prediction["confidence"])
-                mask = np.where(np.array(img) == 0, 1, 0)
-                proba: np.ndarray = np.array([255 - p, 0, p])
-                color = mask * proba[np.newaxis, np.newaxis, :]
-                white_mask = 255 * (1 - mask)
-                img = color + white_mask
-
-            # Write to response page
-            response[ymin:ymax, xmin:xmax, :] = np.array(img)
-
-    return response
-
-
 def draw_boxes(boxes: np.ndarray, image: np.ndarray, color: Optional[Tuple[int, int, int]] = None, **kwargs) -> None:
     """Draw an array of relative straight boxes on an image
 

doctr/version.py

@@ -1 +1 @@
-__version__ = 'v0.8.1'
+__version__ = 'v0.10.0'

{python_doctr-0.8.1.dist-info → python_doctr-0.10.0.dist-info}/METADATA

@@ -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
@@ -219,87 +219,93 @@ 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: 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: 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: tqdm >=4.30.0
+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,>=4.11.0
+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: Pillow>=9.2.0
+Requires-Dist: defusedxml>=0.7.0
+Requires-Dist: anyascii>=0.3.2
+Requires-Dist: tqdm>=4.30.0
+Provides-Extra: contrib
+Requires-Dist: onnxruntime>=1.11.0; extra == "contrib"
 Provides-Extra: dev
-Requires-Dist: tensorflow <2.16.0,>=2.11.0 ; 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: onnx <3.0.0,>=1.12.0 ; 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: 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: 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'
-Requires-Dist: furo >=2022.3.4 ; 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,>=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: 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>=0.3.0; 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<0.22; 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"
+Requires-Dist: furo>=2022.3.4; extra == "dev"
 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: 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'
+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<0.22; 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: html
+Requires-Dist: weasyprint>=55.0; extra == "html"
 Provides-Extra: quality
-Requires-Dist: ruff >=0.1.5 ; extra == 'quality'
-Requires-Dist: mypy >=0.812 ; extra == 'quality'
-Requires-Dist: pre-commit >=2.17.0 ; extra == 'quality'
+Requires-Dist: ruff>=0.1.5; extra == "quality"
+Requires-Dist: mypy>=0.812; extra == "quality"
+Requires-Dist: pre-commit>=2.17.0; extra == "quality"
 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'
+Requires-Dist: pytest>=5.3.2; extra == "testing"
+Requires-Dist: coverage[toml]>=4.5.4; 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"
 Provides-Extra: tf
-Requires-Dist: tensorflow <2.16.0,>=2.11.0 ; extra == 'tf'
-Requires-Dist: tf2onnx <2.0.0,>=1.16.0 ; 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: onnx <3.0.0,>=1.12.0 ; 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: viz
+Requires-Dist: matplotlib>=3.1.0; extra == "viz"
+Requires-Dist: mplcursors>=0.3; extra == "viz"
 
 <p align="center">
   <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