### Clone dhSegment Repository

Source: https://dhsegment.readthedocs.io/en/latest/start/demo.html

Clone the dhSegment repository to your local machine to get started.

```bash
git clone https://github.com/dhlab-epfl/dhSegment.git
```

--------------------------------

### get_classes_color_from_file_multilabel

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Gets classes and code labels from a text file for multi-label scenarios.

```APIDOC
## get_classes_color_from_file_multilabel

### Description
Get classes and code labels from txt file for multi-label scenarios.

### Parameters
* **...**: Parameters for the function.
```

--------------------------------

### Define Class RGB Values

Source: https://dhsegment.readthedocs.io/en/latest/intro/intro.html

This snippet shows how to define the RGB values for different classes in a text file. Each line corresponds to a class, starting with class 0.

```text
0 255 0
255 0 0
0 0 255

```

--------------------------------

### Get Annotations Per File

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Retrieves annotations specific to a given file from VIA JSON content.

```APIDOC
## get_annotations_per_file

### Description
From VIA json content, get annotations relative to the given name_file.

### Parameters
- **via_dict** (`dict`) – VIA annotations content (originally json)
- **name_file** (`str`) – the file to look for (it can be a iiif path or a file path)

### Return type
`dict`

### Returns
dict
```

--------------------------------

### Get VIA Attributes

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Extracts attributes from annotated data and returns a list of VIAttribute objects.

```APIDOC
## get_via_attributes

### Description
Gets the attributes of the annotated data and returns a list of VIAttribute.

### Parameters
- **annotation_dict** (`dict`) – json content of the VIA exported file
- **via_version** (`int`) – either 1 or 2 (for VIA v 1.0 or VIA v 2.0)

### Return type
`List`[`VIAttribute`]

### Returns
A list containing VIAttributes
```

--------------------------------

### Download Provided Model

Source: https://dhsegment.readthedocs.io/en/latest/start/demo.html

Download the provided pre-trained model and unzip it into the demo/model directory. This is recommended to skip the time-consuming training process.

```bash
cd demo/
wget https://github.com/dhlab-epfl/dhSegment/releases/download/v0.2/model.zip
unzip model.zip
cd ..
```

--------------------------------

### Run Training Script

Source: https://dhsegment.readthedocs.io/en/latest/start/training.html

Execute the training script using the 'sacred' package by providing a configuration file.

```bash
python train.py with <config.json>
```

--------------------------------

### Download and Unzip Annotated Dataset

Source: https://dhsegment.readthedocs.io/en/latest/start/demo.html

Download the annotated dataset and unzip it into the demo/pages directory. This dataset contains pre-organized folders for training, validation, and testing.

```bash
cd demo/
wget https://github.com/dhlab-epfl/dhSegment/releases/download/v0.2/pages.zip
unzip pages.zip
cd ..
```

--------------------------------

### Download Pretrained ResNet Model

Source: https://dhsegment.readthedocs.io/en/latest/start/demo.html

Download the pre-trained weights for ResNet. This step is only necessary if you plan to train the model from scratch.

```python
cd pretrained_models/
python download_resnet_pretrained_model.py
cd ..
```

--------------------------------

### Run Demo Script

Source: https://dhsegment.readthedocs.io/en/latest/start/demo.html

Execute the main demo script to perform page document extraction. Results will be saved in the demo/processed_images directory.

```python
python demo.py
```

--------------------------------

### TextLine Instance Methods

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Methods for manipulating and serializing TextLine objects.

```APIDOC
## TextLine Instance Methods

### `scale_baseline_points`

Scales the points of the baseline by a factor ratio.

**Parameters**

*   **ratio** (`float`) – factor to rescale the baseline coordinates

### `to_dict`

Converts a TextLine object to a dictionary.

**Parameters**

*   **non_serializable_keys** (`List`[`str`]) – list of keys that can’t be directly serialized and that need some internal serialization

**Returns**

a dictionary with the atributes of the object serialized

### `to_xml`

Converts a TextLine object to a XML structure.

**Parameters**

*   **name_element** – name of the object (optional)

**Returns**

a etree structure
```

--------------------------------

### TrainingParams

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Parameters to configure the training process.

```APIDOC
## TrainingParams

### Description
Parameters to configure the training process.

### Parameters
* **kwargs**: Arbitrary keyword arguments.
```

--------------------------------

### Class Definition File Format

Source: https://dhsegment.readthedocs.io/en/latest/start/training.html

Defines the RGB color codes for each segmentation class in the 'classes.txt' file. Each class must have a unique color.

```text
classes.txt

0 0 0
0 255 0
...


```

--------------------------------

### Training Parameters Class

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Class for configuring the training process.

```APIDOC
## TrainingParams(_**kwargs_)

Parameters to configure the training process.

### Variables

- **n_epochs** (int): Number of epochs for training.
- **evaluate_every_epoch** (int): The model will be evaluated every n epochs.
- **learning_rate** (float): The starting learning rate value.
- **exponential_learning** (bool): Option to use exponential learning rate.
- **batch_size** (int): Size of batch.
- **data_augmentation** (bool): Option to use data augmentation (by default is set to False).
- **data_augmentation_flip_lr** (bool): Option to use image flipping in right-left direction.
- **data_augmentation_flip_ud** (bool): Option to use image flipping in up down direction.
- **data_augmentation_color** (bool): Option to use data augmentation with color.
- **data_augmentation_max_rotation** (float): Maximum angle of rotation (in radians) for data augmentation.
- **data_augmentation_max_scaling** (float): Maximum scale of zooming during data augmentation (range: [0,1]).
- **make_patches** (bool): Option to crop image into patches. This will cut the entire image in several patches.
- **patch_shape** (tuple): Shape of the patches.
- **input_resized_size** (int): Size (in pixel) of the image after resizing. The original ratio is kept. If no resizing is wanted, set it to -1.
- **weights_labels** (list): Weight given to each label. Should be a list of length = number of classes.
- **training_margin** (int): Size of the margin to add to the images. This is particularly useful when training with patches.
- **local_entropy_ratio** (float): 
- **local_entropy_sigma** (float): 
- **focal_loss_gamma** (float): Value of gamma for the focal loss. See paper : https://arxiv.org/abs/1708.02002

### Methods

- `check_params()`: Checks if there is no parameter inconsistency. Returns None.
```

--------------------------------

### Miscellaneous Helpers

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

A collection of general-purpose utility functions for file handling and data manipulation.

```APIDOC
## Miscellaneous helpers

- `parse_json(filename)`: Parses JSON data from a file.
- `dump_json(filename, dict)`: Dumps a dictionary to a JSON file.
- `load_pickle(filename)`: Loads data from a pickle file.
- `dump_pickle(filename, obj)`: Dumps an object to a pickle file.
- `hash_dict(params)`: Hashes a dictionary.
```

--------------------------------

### TextLine Class Methods

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Methods for creating TextLine objects from various formats.

```APIDOC
## TextLine Class Methods

### `from_array`

Creates a TextLine object from array inputs.

### `from_dict`

Creates a TextLine object from a serialized dictionary.

**Parameters**

*   **dictionary** (`dict`) – serialized dictionary

**Returns**

`TextLine`

### `from_xml`

Creates a TextLine object from an XML structure.

**Parameters**

*   **etree_element** (`Element`) – a xml etree

**Returns**

a dictionary with keys ‘id’ and ‘coords’
```

--------------------------------

### TextRegion Instance Methods

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Methods for manipulating and serializing TextRegion objects.

```APIDOC
## TextRegion Instance Methods

### `sort_text_lines`

Sorts `TextLine` from top to bottom according to their mean y coordinate (centroid).

**Parameters**

*   **top_to_bottom** (`bool`) – order lines from top to bottom of image, default=True

**Returns**

`None`

### `to_dict`

Converts a TextRegion object to a dictionary.

**Parameters**

*   **non_serializable_keys** (`List`[`str`]) – list of keys that can’t be directly serialized and that need some internal serialization

**Returns**

a dictionary with the atributes of the object serialized

### `to_xml`

Converts a TextRegion object to a XML structure.

**Parameters**

*   **name_element** – name of the object (optional)

**Returns**

a etree structure
```

--------------------------------

### WorkingItem Class

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

A container for annotated images within the VIA format.

```APIDOC
## WorkingItem Class

### Description
A container for annotated images. This class holds information about an image and its associated annotations in the VIA format.

### Parameters
- **collection** (str) - name of the collection
- **image_name** (str) - name of the image
- **original_x** (int) - original image x size (width)
- **original_y** (int) - original image y size (height)
- **reduced_x** (int) - resized x size
- **reduced_y** (int) - resized y size
- **iiif** (str) - iiif url
- **annotations** (dict) - VIA ‘region_attributes’

### Properties
- **collection** (str) - Alias for field number 0
- **image_name** (str) - Alias for field number 1
- **original_x** (int) - Alias for field number 2
- **original_y** (int) - Alias for field number 3
- **reduced_x** (int) - Alias for field number 4
- **reduced_y** (int) - Alias for field number 5
- **iiif** (str) - Alias for field number 6
- **annotations** (dict) - Alias for field number 7
```

--------------------------------

### dh_segment.io.serving_input_image

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Generates serving input for image-based input.

```APIDOC
## dh_segment.io.serving_input_image

### Description
Generates serving input for image-based input.

### Returns
fn - Serving input function.
```

--------------------------------

### Create and Activate Virtual Environment

Source: https://dhsegment.readthedocs.io/en/latest/start/install.html

Use these commands to create a new conda environment named 'dh_segment' with Python 3.6 and then activate it. This isolates project dependencies.

```bash
conda create -n dh_segment python=3.6
source activate dh_segment

```

--------------------------------

### create_via_annotation_single_image

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Creates a VIA annotation dictionary for a single image.

```APIDOC
## create_via_annotation_single_image

### Description
Returns a dictionary item {key: annotation} in VIA format to further export to .json file.

### Parameters
- **img_filename** (str) - path to the image
- **via_regions** (List[dict]) - regions in VIA format (output from `create_via_region_from_coordinates`)
- **file_attributes** (Optional[dict]) - file attributes (usually None)

### Return Type
`Dict[str, dict]`

Returns a dictionary item with key and annotations in VIA format.
```

--------------------------------

### CSV Input Data Format

Source: https://dhsegment.readthedocs.io/en/latest/start/training.html

Use this format when providing training data via a CSV file, listing pairs of original and label image filenames.

```text
mypath/myfolder/original_image_filename1,mypath/myfolder/label_image_filename1
mypath/myfolder/original_image_filename2,mypath/myfolder/label_image_filename2


```

--------------------------------

### dh_segment.io.load_and_resize_image

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Loads an image from its filename and resizes it.

```APIDOC
## dh_segment.io.load_and_resize_image

### Description
Loads an image from its filename and resizes it to the desired output size.

### Parameters
* **filename** (str) - Required - String tensor.
* **channels** (int) - Required - Number of channels for the decoded image.
* **size** (Optional[int]) - Optional - Number of desired pixels in the resized image, tf.Tensor or int (None for no resizing).
* **interpolation** (str) - Optional - Interpolation method.
* **return_original_shape** (bool) - Optional - Returns the original shape of the image before resizing if this flag is True.

### Returns
tensorflow.Tensor - The loaded and resized image.
```

--------------------------------

### Image and Class Conversion Utilities

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Functions for converting between image representations and class labels.

```APIDOC
## Image and Class Conversion Utilities

- `label_image_to_class(label_image, classes_file)`: Converts a label image to class format. Returns tensorflow.Tensor.
- `class_to_label_image(class_label, classes_file)`: Converts class labels to a label image. Returns tensorflow.Tensor.
- `multilabel_image_to_class(label_image, classes_file)`:
    Combines image annotations with classes info of the txt file to create the input label for the training.

    Parameters:
      - **label_image** (tensorflow.Tensor): Annotated image [H,W,Ch] or [B,H,W,Ch] (Ch = color channels).
      - **classes_file** (str): The filename of the txt file containing the class info.

    Returns:
      tensorflow.Tensor: [H,W,Cl] or [B,H,W,Cl] (Cl = number of classes).
- `multiclass_to_label_image(class_label_tensor, classes_file)`: Converts multi-class labels to a label image. Returns tensorflow.Tensor.
- `get_classes_color_from_file(classes_file)`: Retrieves class colors from a file. Returns ndarray.
- `get_n_classes_from_file(classes_file)`: Retrieves the number of classes from a file. Returns int.
- `get_classes_color_from_file_multilabel(classes_file)`:
    Get classes and code labels from txt file. This function deals with the case of elements with multiple labels.

    Parameters:
      - **classes_file** (str): File containing the classes (usually named _classes.txt_).

    Returns:
      Tuple[ndarray, array]: A tuple containing class colors and code labels.
```

--------------------------------

### TextRegion Class Methods

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Methods for creating TextRegion objects from various formats.

```APIDOC
## TextRegion Class Methods

### `from_dict`

Creates a TextRegion object from a serialized dictionary.

**Parameters**

*   **dictionary** (`dict`) – serialized dictionary

**Returns**

`TextRegion`

### `from_xml`

Creates a TextRegion object from an XML structure.

**Parameters**

*   **etree_element** – a xml etree

**Returns**

a dictionary with keys ‘id’ and ‘coords’
```

--------------------------------

### dh_segment.io.serving_input_filename

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Generates serving input for filename-based input.

```APIDOC
## dh_segment.io.serving_input_filename

### Description
Generates serving input for filename-based input.

### Parameters
* **resized_size** (int) - The desired size for resizing.

### Returns
fn - Serving input function.
```

--------------------------------

### Load Annotation Data

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Loads the content of VIA annotation files.

```APIDOC
## load_annotation_data

### Description
Load the content of via annotation files.

### Parameters
- **via_data_filename** (`str`) – via annotations json file
- **only_img_annotations** (`bool`) – load only the images annotations (‘_via_img_metadata’ field)
- **via_version** (`int`) – 

### Return type
`dict`

### Returns
the content of json file containing the region annotated
```

--------------------------------

### multilabel_image_to_class

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Combines image annotations with classes info to create the input label for training.

```APIDOC
## multilabel_image_to_class

### Description
Combines image annotations with classes info of the txt file to create the input label for the training.

### Parameters
* **label_image**: The input label image.
* **...**: Additional parameters.
```

--------------------------------

### Metrics

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Utility class for evaluation metrics.

```APIDOC
## Metrics

### Description
Utility class for evaluation metrics.
```

--------------------------------

### BaseElement

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Base class for page elements. This is an abstract class.

```APIDOC
## BaseElement

### Description
Base page element class. (Abstract)

### Class Methods
* **check_tag**(_tag_)
* **full_tag**() - Returns the full tag of the element.

### Attributes
* **tag** (str, optional) - The tag associated with the element.
```

--------------------------------

### local_entropy

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Calculates the local entropy of a binary image.

```APIDOC
## local_entropy

### Parameters
* **tf_binary_img** (tensorflow.Tensor) - The input binary image tensor.
* **sigma** (float) - The sigma value for the Gaussian kernel, defaults to 3.

### Returns
tensorflow.Tensor - The calculated local entropy.
```

--------------------------------

### Model Parameter Classes

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Classes for defining model-specific parameters.

```APIDOC
## Model Parameter Classes

### VGG16ModelParams

Parameters for the VGG16 model.

- `CORRECTED_VERSION`: (None)
- `INTERMEDIATE_CONV`: [[(256, 3)]]
- `PRETRAINED_MODEL_FILE`: 'pretrained_models/vgg_16.ckpt'
- `SELECTED_LAYERS_UPSCALING`: [True, True, True, True, False, False]
- `UPSCALE_PARAMS`: [[(32, 3)], [(64, 3)], [(128, 3)], [(256, 3)], [(512, 3)], [(512, 3)]]

### ResNetModelParams

Parameters for the ResNet model.

- `CORRECT_VERSION`: False
- `INTERMEDIATE_CONV`: None
- `PRETRAINED_MODEL_FILE`: 'pretrained_models/resnet_v1_50.ckpt'
- `SELECTED_LAYERS_UPSCALING`: [True, True, True, True, True]
- `UPSCALE_PARAMS`: [(32, 0), (64, 0), (128, 0), (256, 0), (512, 0)]

### UNetModelParams

Parameters for the UNet model.

- `CORRECT_VERSION`: False
- `INTERMEDIATE_CONV`: None
- `PRETRAINED_MODEL_FILE`: None
- `SELECTED_LAYERS_UPSCALING`: None
- `UPSCALE_PARAMS`: None

### ModelParams(_**kwargs_)

Parameters related to the model.

- `check_params()`: Checks model parameters.
```

--------------------------------

### get_n_classes_from_file_multilabel

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Retrieves the number of classes for multi-label scenarios from a file.

```APIDOC
## get_n_classes_from_file_multilabel

### Description
Retrieves the number of classes for multi-label scenarios from a file.

### Parameters
* **classes_file**: Path to the file containing class information.

### Returns
* **int**: The number of classes.
```

--------------------------------

### ModelParams

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Parameters related to the model configuration.

```APIDOC
## ModelParams

### Description
Parameters related to the model configuration.

### Parameters
* **kwargs**: Arbitrary keyword arguments.
```

--------------------------------

### Create Masks from VIA Annotations

Source: https://dhsegment.readthedocs.io/en/latest/start/annotating.html

This script processes VGG Image Annotator (VIA) JSON annotations to generate individual image masks for each annotated region and attribute option. Ensure VIA annotations favor 'dropdown', 'checkbox', or 'radio' attribute types for easier parsing.

```python
from dh_segment.io import via

collection = 'mycollection'
annotation_file = 'via_sample.json'
masks_dir = '/home/project/generated_masks'
images_dir = './my_images'

# Load all the data in the annotation file
# (the file may be an exported project or an export of the annotations)
via_data = via.load_annotation_data(annotation_file)

# In the case of an exported project file, you can set ``only_img_annotations=True``
# to get only the image annotations
via_annotations = via.load_annotation_data(annotation_file, only_img_annotations=True)

# Collect the annotated regions
working_items = via.collect_working_items(via_annotations, collection, images_dir)

# Collect the attributes and options
if '_via_attributes' in via_data.keys():
    list_attributes = via.parse_via_attributes(via_data['_via_attributes'])
else:
    list_attributes = via.get_via_attributes(via_annotations)

# Create one mask per option per attribute
via.create_masks(masks_dir, working_items, list_attributes, collection)

```

--------------------------------

### Tensorflow serving functions

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Functions for creating serving input receivers for TensorFlow models.

```APIDOC
## serving_input_filename

### Description
Serving input function based on filename.

### Parameters
- **resized_size**: Type and description not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

```APIDOC
## serving_input_image

### Description
Serving input function based on image data.

### Parameters
None specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

--------------------------------

### thresholding

Source: https://dhsegment.readthedocs.io/en/latest/reference/post_processing.html

Computes the binary mask of the detected Page from the probabilities output by network.

```APIDOC
## thresholding

### Description
Computes the binary mask of the detected Page from the probabilities output by network.

### Method
`thresholding(probs, threshold=-1)`

### Parameters
#### Path Parameters
- **probs** (ndarray) - Required - array in range [0, 1] of shape HxWx2
- **threshold** (float) - Optional - threshold between [0 and 1], if negative Otsu’s adaptive threshold will be used

### Return type
`ndarray`

### Returns
binary mask
```

--------------------------------

### cleaning_binary

Source: https://dhsegment.readthedocs.io/en/latest/reference/post_processing.html

Uses mathematical morphology to clean and remove small elements from binary images.

```APIDOC
## cleaning_binary

### Description
Uses mathematical morphology to clean and remove small elements from binary images.

### Method
`cleaning_binary(mask, kernel_size=5)`

### Parameters
#### Path Parameters
- **mask** (ndarray) - Required - the binary image to clean
- **kernel_size** (int) - Optional - size of the kernel

### Return type
`ndarray`

### Returns
the cleaned mask
```

--------------------------------

### Utility Functions

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Helper functions for processing XML and JSON data.

```APIDOC
## Utility Functions

### `get_unique_tags_from_xml_text_regions`

Get a list of all the values of labels/tags from XML text regions.

**Parameters**

*   **xml_filename** (`str`) – filename of the xml file
*   **tag_pattern** (`str`) – regular expression pattern to look for in TextRegion.custom_attribute

**Returns**

### `json_serialize`

Serialize a dictionary in order to export it.

**Parameters**

*   **dict_to_serialize** (`dict`) – dictionary to serialize
*   **non_serializable_keys** (`List`[`str`]) – keys that are not directly seriazable sucha as python objects

**Returns**

`dict` – the serialized dictionnary

### `parse_file`

Parses the files to create the corresponding `Page` object. The files can be a .xml or a .json.

**Parameters**

*   **filename** (`str`)
```

--------------------------------

### create_via_region_from_coordinates

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Formats coordinates into a VIA region dictionary.

```APIDOC
## create_via_region_from_coordinates

### Description
Formats coordinates to a VIA region (dict).

### Parameters
- **coordinates** (array) - (N, 2) coordinates (x, y)
- **region_attributes** (dict) - dictionary with keys : name of labels, values : values of labels
- **type_region** (str) - via region annotation type (‘rect’, ‘polygon’)

### Return Type
`dict`

Returns a region in VIA style (dict/json).
```

--------------------------------

### predict_with_tiles Method

Source: https://dhsegment.readthedocs.io/en/latest/reference/inference.html

Performs prediction on an image by dividing it into tiles. This is useful for large images that cannot be processed in memory at once.

```APIDOC
## LoadedModel.predict_with_tiles

### Description
Performs prediction on an image by dividing it into tiles.

### Parameters
* **filename** (str) - The path to the image file.
* **resized_size** (tuple, optional) - The target size to resize the image to before tiling. If None, the original size is used.
* **tile_size** (int, optional) - The size of each tile. Defaults to 500.
* **min_overlap** (float, optional) - The minimum overlap between adjacent tiles. Defaults to 0.2.
* **linear_interpolation** (bool, optional) - Whether to use linear interpolation when combining tile predictions. Defaults to True.
```

--------------------------------

### Point.cv2_to_point_list

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Converts an opencv-formatted set of coordinates to a list of Point.

```APIDOC
## Point.cv2_to_point_list

### Description
Converts an opencv-formatted set of coordinates to a list of Point.

### Parameters

#### Parameters
- **cv2_array** (ndarray) - opencv-formatted set of coordinates, shape (N,1,2)

### Returns
List[Point]
```

--------------------------------

### PAGE XML and JSON import / export

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Classes and functions for parsing and serializing PAGE XML and JSON data.

```APIDOC
## PAGE.Point

### Description
Represents a point with (x,y) coordinates.

### Parameters
- **y**: Type and description not specified in source.
- **x**: Type and description not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.Text

### Description
Represents a text entity, typically produced by a transcription system.

### Parameters
- **text_equiv**: Type and description not specified in source.
- **alternatives**: Type and description not specified in source.
- **score**: Type and description not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.Border

### Description
Represents a region containing the page border.

### Parameters
- **coords**: Type and description not specified in source.
- **id**: Type and description not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.TextRegion

### Description
Represents a region containing text lines.

### Parameters
- **id**: Type and description not specified in source.
- **coords**: Type and description not specified in source.
- **text_lines**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.TextLine

### Description
Represents a region corresponding to a text line.

### Parameters
- **id**: Type and description not specified in source.
- **coords**: Type and description not specified in source.
- **baseline**: Type and description not specified in source.
- **text**: Type and description not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.GraphicRegion

### Description
Represents a region containing simple graphics.

### Parameters
- **id**: Type and description not specified in source.
- **coords**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.TableRegion

### Description
Represents a region containing tabular data.

### Parameters
- **id**: Type and description not specified in source.
- **coords**: Type and description not specified in source.
- **rows**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.SeparatorRegion

### Description
Represents lines separating columns or paragraphs.

### Parameters
- **id**: Type and description not specified in source.
- **coords**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.GroupSegment

### Description
Represents a set of regions that form a larger region (group).

### Parameters
- **id**: Type and description not specified in source.
- **coords**: Type and description not specified in source.
- **segment_ids**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.Metadata

### Description
Represents metadata information.

### Parameters
- **creator**: Type and description not specified in source.
- **created**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.Page

### Description
Represents a class following the PAGE-XML object structure.

### Parameters
- **kwargs**: Type and description not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.BaseElement

### Description
Base class for PAGE elements.

### Parameters
None specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.Region

### Description
Base class for PAGE regions.

### Parameters
- **id**: Type and description not specified in source.
- **coords**: Type and description not specified in source.
- **custom_attribute**: Type and description not specified in source.

### Method
Not applicable (Python class)

### Endpoint
Not applicable (Python class)
```

```APIDOC
## PAGE.parse_file

### Description
Parses a file to create the corresponding `Page` object.

### Parameters
- **filename**: Type and description not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

```APIDOC
## PAGE.json_serialize

### Description
Serializes a dictionary for export.

### Parameters
- **dict_to_serialize**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

--------------------------------

### Point Conversion

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Provides a class method to convert a list of Point objects into a list of coordinates.

```APIDOC
## _classmethod_`point_to_list`(_points_)

### Description
Converts a list of Point to a list of coordinates.

### Parameters
#### Path Parameters
- **points** (List[Point]) - Required - list of Points

### Returns
- **list** - list of shape (N,2)
```

--------------------------------

### Point.list_from_xml

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Converts a PAGEXML-formatted set of coordinates to a list of Point.

```APIDOC
## Point.list_from_xml

### Description
Converts a PAGEXML-formatted set of coordinates to a list of Point.

### Parameters

#### Parameters
- **etree_elem** (Element) - etree XML element containing a set of coordinates

### Returns
List[Point]
```

--------------------------------

### inference_u_net

Source: https://dhsegment.readthedocs.io/en/latest/reference/network.html

Performs inference using the U-Net model architecture.

```APIDOC
## inference_u_net

### Description
Performs inference using the U-Net model architecture.

### Parameters
- **images** (type) - Description
- **params** (type) - Description
- **num_classes** (type) - Description
- **use_batch_norm** (bool) - Optional - Whether to use batch normalization.
- **weight_decay** (float) - Optional - The weight decay factor.
- **is_training** (bool) - Optional - Whether the model is in training mode.

### Return Type
tensorflow.Tensor
```

--------------------------------

### create_masks

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Creates binary masks for annotations from VIA data.

```APIDOC
## create_masks

### Description
For each annotation, create a corresponding binary mask and resize it (h = 2000). Only valid for VIA 2.0. Several annotations of the same class on the same image produce one image with several masks.

### Parameters
- **masks_dir** (str) - where to output the masks
- **working_items** (List[WorkingItem]) - infos to work with
- **via_attributes** (List[VIAttribute]) - VIAttributes computed by `get_via_attributes` function.
- **collection** (str) - name of the collection
- **contours_only** (bool) - creates the binary masks only for the contours of the object (thickness of contours : 20 px)

### Return Type
`dict`

Returns annotation_summary, a dictionary containing a list of labels per image.
```

--------------------------------

### Point.list_to_point

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Converts a list of coordinates to a list of Point.

```APIDOC
## Point.list_to_point

### Description
Converts a list of coordinates to a list of Point.

### Parameters

#### Parameters
- **list_coords** (list) - list of coordinates, shape (N, 2)

### Returns
List[Point]
```

--------------------------------

### get_classes_color_from_file

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Retrieves class colors from a file.

```APIDOC
## get_classes_color_from_file

### Description
Retrieves class colors from a file.

### Parameters
* **classes_file**: Path to the file containing class information.

### Returns
* **ndarray**: An array containing class colors.
```

--------------------------------

### Page

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Represents the entire page information, including image details, text regions, graphic regions, and metadata. Can be exported to PAGE-XML or JSON.

```APIDOC
## Page

### Description
Class following PAGE-XML object. This class is used to represent the information of the processed image. It is possible to export this info as PAGE-XML or JSON format.

### Attributes
* **image_filename** (str) - Filename of the image.
* **image_width** (int) - Width of the original image.
* **image_height** (int) - Height of the original image.
* **text_regions** (list) - List of TextRegion objects.
* **graphic_regions** (list) - List of GraphicRegion objects.
* **page_border** (Border) - Border object defining the page boundaries.
* **separator_regions** (list) - List of SeparatorRegion objects.
* **table_regions** (list) - List of TableRegion objects.
* **metadata** (Metadata) - Metadata object containing information about the image and process.
* **line_groups** (list) - List of GroupSegment objects forming lines.
```

--------------------------------

### Metadata

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Stores metadata information about the image processing, including creator, creation time, and comments.

```APIDOC
## Metadata

### Description
Metadata information.

### Class Methods
* **from_dict**(_dictionary_) - Creates a Metadata object from a dictionary.
* **from_xml**(_e_) - Creates a Metadata object from an XML element.

### Methods
* **to_dict**() - Converts the Metadata object to a dictionary.
* **to_xml**() - Converts the Metadata object to an XML Element.

### Attributes
* **creator** - Name of the process or person that created the exported file.
* **created** - Time of creation of the file.
* **last_change** - Time of last modification of the file.
* **comments** - Comments on the process.
* **tag** - 'Metadata'
```

--------------------------------

### Input functions for tf.Estimator

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Provides input functions and data augmentation utilities for use with TensorFlow Estimators.

```APIDOC
## input_fn

### Description
Input function for TensorFlow Estimators.

### Parameters
- **input_data**: Type and description not specified in source.
- **params**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

```APIDOC
## data_augmentation_fn

### Description
Applies data augmentation to both input images and label images.

### Parameters
- **input_image**: Type and description not specified in source.
- **label_image**: Type and description not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

```APIDOC
## extract_patches_fn

### Description
Cuts a given image into patches.

### Parameters
- **image**: Type and description not specified in source.
- **patch_shape**: Type and description not specified in source.
- **offsets**: Type and description not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

```APIDOC
## rotate_crop

### Description
Rotates and crops the images.

### Parameters
- **image**: Type and description not specified in source.
- **rotation**: Type and description not specified in source.
- **crop**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

```APIDOC
## resize_image

### Description
Resizes the image to a specified size.

### Parameters
- **image**: Type and description not specified in source.
- **size**: Type and description not specified in source.
- **interpolation**: Type and description not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

```APIDOC
## load_and_resize_image

### Description
Loads an image from its filename and resizes it to the desired output size.

### Parameters
- **filename**: Type and description not specified in source.
- **channels**: Type and description not specified in source.
- **...**: Additional parameters not specified in source.

### Method
Not applicable (Python function)

### Endpoint
Not applicable (Python function)
```

--------------------------------

### write_to_file

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Export Page object to json or page-xml format. Will assume the format based on the extension of the filename, if there is no extension will export as an xml file.

```APIDOC
## write_to_file

### Description
Export Page object to json or page-xml format. Will assume the format based on the extension of the filename, if there is no extension will export as an xml file.

### Parameters

#### Parameters
- **filename** (str) - filename of the file to be exported
- **creator_name** (str) - name of the creator (process or person) creating the file
- **comments** (str) - optionnal comment to add to the metadata of the file.

### Return Type
None
```

--------------------------------

### draw_baselines

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Draws the TextLines' baselines on the image canvas. Coordinates can be autiscaled.

```APIDOC
## draw_baselines

### Description
Given an image, draws the TextLines.baselines.

### Parameters
* **img_canvas** (`ndarray`) – 3 channel image in which the region will be drawn. The image is modified inplace.
* **color** (`Tuple`[`int`, `int`, `int`]) – (R, G, B) value color
* **thickness** (`int`) – the thickness of the line
* **endpoint_radius** (`int`) – the radius of the endpoints of line s(first and last coordinates of line)
* **autoscale** (`bool`) – whether to scale the coordinates to the size of img_canvas. If True, it will use the dimensions provided in Page.image_width and Page.image_height to compute the scaling ratio
```

--------------------------------

### LoadedModel Class

Source: https://dhsegment.readthedocs.io/en/latest/reference/inference.html

Loads an exported dhSegment model. This class is used to initialize a model for inference.

```APIDOC
## class dh_segment.inference.LoadedModel

### Description
Loads an exported dhSegment model.

### Parameters
* **model_base_dir** (str) - The model directory containing saved_model.{pb|pbtxt}. If not provided, it is assumed to be a TF exporter directory, and the latest export directory will be automatically selected.
* **predict_mode** (str, optional) - Defines the input/output format of the prediction output. Defaults to 'filename'.
* **num_parallel_predictions** (int, optional) - Limits the number of concurrent calls to predict to avoid Out-Of-Memory issues if predicting on GPU. Defaults to 2.
```

--------------------------------

### VIAttribute Class

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

A container for VIA attributes, representing metadata for annotations.

```APIDOC
## VIAttribute Class

### Description
A container for VIA attributes. This class holds information about annotation attributes like name, type, and options.

### Parameters
- **name** (str) - The name of attribute
- **type** (str) - The type of the annotation (dropdown, markbox, …)
- **options** (list) - The options / labels possible for this attribute.

### Properties
- **name** (str) - Alias for field number 0
- **type** (str) - Alias for field number 1
- **options** (list) - Alias for field number 2
```

--------------------------------

### Point.list_point_to_string

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Converts a list of Point to a string ‘x,y’.

```APIDOC
## Point.list_point_to_string

### Description
Converts a list of Point to a string ‘x,y’.

### Parameters

#### Parameters
- **list_points** (List[Point]) - list of coordinates with Point format

### Returns
a string with the coordinates
```

--------------------------------

### multiclass_to_label_image

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Converts multi-class labels to a label image.

```APIDOC
## multiclass_to_label_image

### Description
Converts multi-class labels to a label image.

### Returns
* **tensorflow.Tensor**: The resulting label image.
```

--------------------------------

### get_n_classes_from_file

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Retrieves the number of classes from a file.

```APIDOC
## get_n_classes_from_file

### Description
Retrieves the number of classes from a file.

### Parameters
* **classes_file**: Path to the file containing class information.

### Returns
* **int**: The number of classes.
```

--------------------------------

### Multilabel Class Definition

Source: https://dhsegment.readthedocs.io/en/latest/start/training.html

For multilabel classification, extend 'classes.txt' with an attribution code indicating which classes are assigned to each color. The attribution code has a length equal to the number of classes.

```text
classes.txt

0 0 0 0 0 0
0 255 0 1 0 0
255 0 0 0 1 0
0 0 255 0 0 1
128 128 128 1 1 0
0 255 255 0 1 1


```

--------------------------------

### export_annotation_dict

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Exports an annotation dictionary to a file.

```APIDOC
## export_annotation_dict

### Description
Exports an annotation dictionary to a specified file.

### Parameters
- **annotation_dict** - The dictionary containing annotations to export.
- **filename** (str) - The name of the file to save the annotations to.
```

--------------------------------

### label_image_to_class

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Converts a label image to class labels.

```APIDOC
## label_image_to_class

### Description
Converts a label image to class labels.

### Parameters
* **label_image**: The input label image.
* **classes_file**: Path to the file containing class information.

### Returns
* **tensorflow.Tensor**: The resulting class labels.
```

--------------------------------

### Parse VIA Attributes

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Parses a VIA attribute dictionary into a list of VIAttribute instances.

```APIDOC
## parse_via_attributes

### Description
Parses the VIA attribute dictionary and returns a list of VIAttribute instances.

### Parameters
- **via_attributes** (`dict`) – attributes from VIA annotation (‘_via_attributes’ field)

### Return type
`List`[`VIAttribute`]

### Returns
list of `VIAttribute`
```

--------------------------------

### Point.array_to_point

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Converts an np.array to a list of Point.

```APIDOC
## Point.array_to_point

### Description
Converts an np.array to a list of Point.

### Parameters

#### Parameters
- **array** (ndarray) - an array of coordinates. Must be of shape (N, 2)

### Returns
list of Point
```

--------------------------------

### collect_working_items

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Collects all WorkingItem objects from VIA annotations.

```APIDOC
## collect_working_items

### Description
Given VIA annotation input, collect all info on WorkingItem object. This function will take care of separating images from local files and images from IIIF urls.

### Parameters
- **via_annotations** (dict) - via annotations (‘regions’ field)
- **collection_name** (str) - name of the collection
- **images_dir** (Optional[str]) - directory where to find the images
- **via_version** (int) - version of the VIA tool used to produce the annotations (1 or 2)

### Return Type
`List[WorkingItem]`

Returns a list of WorkingItem objects.
```

--------------------------------

### load_pickle

Source: https://dhsegment.readthedocs.io/en/latest/reference/utils.html

Loads a Python object from a pickle file. This function is useful for deserializing objects that were previously saved using Python's pickle module.

```APIDOC
## load_pickle(filename)

### Description
Loads and deserializes a Python object from a pickle file specified by `filename`.

### Parameters
#### Path Parameters
- **filename** (string) - Required - The path to the pickle file to be loaded.
```

--------------------------------

### draw_text_regions

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Given an image, draws the TextRegions, either fills it (fill=True) or draws the contours (fill=False). The image is modified in-place.

```APIDOC
## draw_text_regions

### Description
Given an image, draws the TextRegions, either fills it (fill=True) or draws the contours (fill=False). The image is modified in-place.

### Parameters

#### Parameters
- **img_canvas** (ndarray) - 3 channel image in which the region will be drawn. The image is modified inplace.
- **color** (Tuple[int, int, int]) - (R, G, B) value color
- **fill** (bool) - either to fill the region (True) of only draw the external contours (False)
- **thickness** (int) - in case fill=True the thickness of the line
- **autoscale** (bool) - whether to scale the coordinates to the size of img_canvas. If True, it will use the dimensions provided in Page.image_width and Page.image_height to compute the scaling ratio
```

--------------------------------

### inference_vgg16

Source: https://dhsegment.readthedocs.io/en/latest/reference/network.html

Performs inference using the VGG16 model architecture.

```APIDOC
## inference_vgg16

### Description
Performs inference using the VGG16 model architecture.

### Parameters
- **images** (type) - Description
- **params** (type) - Description
- **num_classes** (type) - Description
- **use_batch_norm** (bool) - Optional - Whether to use batch normalization.
- **weight_decay** (float) - Optional - The weight decay factor.
- **is_training** (bool) - Optional - Whether the model is in training mode.

### Return Type
tensorflow.Tensor
```

--------------------------------

### Export Annotations to JSON

Source: https://dhsegment.readthedocs.io/en/latest/reference/io.html

Exports VIA annotations to a JSON file.

```APIDOC
## export_annotations_to_json

### Description
Exports the annotations to a json file.

### Parameters
- **annotation_dict** (`dict`) – VIA annotations
- **filename** (`str`) – filename to export the data (json file)

### Return type
`None`
```