Tomviz

https://tomviz.readthedocs.io/en/latest
Admin
Tomviz is an open-source, cross-platform application designed for the acquisition, processing, 3D...
Tokens:30,972
Snippets:386
Trust Score:9.6
Update:1 month ago
Show doc for...
Tokens:
Raw
### Install tomviz-pipeline package

Source: https://tomviz.readthedocs.io/en/latest/pipelines.html

Install the pipeline runner from the local repository source.

```bash
git clone --recursive https://github.com/openchemistry/tomviz
cd tomviz/tomviz/python
pip install .
```

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

### Install the Acquisition Server

Source: https://tomviz.readthedocs.io/en/latest/acquisition.html

Commands to clone the repository and set up the Python virtual environment with necessary dependencies.

```bash
git clone --recursive git://github.com/openchemistry/tomviz.git
cd tomviz/acquisition
python -m venv tomviz-acq
source tomviz-acq/bin/activate
pip install -e .
pip install -e .[tiff]
pip install -e .[test]
```

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

### Start the Acquisition Server

Source: https://tomviz.readthedocs.io/en/latest/acquisition.html

Command to launch the passive acquisition server process.

```bash
source tomviz-acq/bin/activate
tomviz-acquisition -a tomviz.acquisition.vendors.passive.PassiveWatchSource
```

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

### Setup External Python Environment

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Commands to create a Conda environment and install the required tomviz-pipeline package.

```bash
conda create -n my_transform_env python=3.10
conda activate my_transform_env
conda install -c conda-forge tomviz-pipeline
pip install torch  # or any other packages your transform needs
```

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

### Start a Test Sequence

Source: https://tomviz.readthedocs.io/en/latest/acquisition.html

Command to simulate data acquisition by writing images from a stack to a directory at a specified interval.

```bash
tomviz-tiltseries-writer -p /tmp/test -d 5 -t tiff
```

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

### Install tomviz-pipeline via conda

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Installs the pipeline execution package from the conda-forge channel.

```bash
conda install -c conda-forge tomviz-pipeline
```

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

### Accessing Dataset in Operator

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/dataset.html

Example of how the Dataset object is provided as the first argument to the transform function in a tomviz operator.

```python
def transform(dataset, rotation_angle=90.0, rotation_axis=0):
    # ...

    array = dataset.active_scalars

    # ...
```

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

### Create a data source

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Initialize a new data source from a file and append it to the application's pipelines.

```python
>>> from tomviz import state
>>> ds = state.DataSource(
            fileNames=['SMALL_PtCu_NanoParticle.tif']
    )
>>> p = state.Pipeline(ds)
>>> state.pipelines.append(p)
```

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

### Taking a screenshot

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Load a state file, execute the pipeline, and save the active view as an image file.

```python
>>> from tomviz import state
>>> state.load('test.tvsm')
>>> state.pipelines[0].execute()
>>> view = state.active_view
>>> view.save_screenshot('screen.png')
```

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

### Download SAM 2 Model Checkpoint

Source: https://tomviz.readthedocs.io/en/latest/ml_segmentation.html

Commands to create the checkpoint directory and download the base plus model weights.

```bash
mkdir -p ~/.tomviz/sam2/checkpoints && cd ~/.tomviz/sam2/checkpoints
curl -O https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_base_plus.pt
```

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

### Create a Spreadsheet with tomviz.utils

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/utils.html

Demonstrates the usage of make_spreadsheet to format data into a dictionary containing a vtkTable object.

```python
spreadsheet = utils.make_spreadsheet(column_names, table)

return {
    'table_data': spreadsheet,
}
```

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

### Configure output format

Source: https://tomviz.readthedocs.io/en/latest/pipelines.html

Specify the output format to bundle pipeline state or generate individual port files.

```bash
tomviz-pipeline -s pipeline.tvsm -o results/ --output-format state
```

```python
run("pipeline.tvsm", "results/", output_format="state")
```

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

### Execute a pipeline

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Load a state file and trigger the execution of the first pipeline in the state.

```python
>>> from tomviz import state
>>> state.load('test.tvsm')
>>> state.pipelines[0].execute()
```

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

### Create SAM 3 Conda Environment

Source: https://tomviz.readthedocs.io/en/latest/ml_segmentation.html

Initializes the required environment using the provided YAML configuration file.

```bash
conda env create -f sam3-tomviz-cuda.yml
```

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

### Execute a pipeline

Source: https://tomviz.readthedocs.io/en/latest/pipelines.html

Run a saved pipeline state file against pinned inputs.

```bash
tomviz-pipeline -s pipeline.tvsm -o results/
```

```python
from tomviz.pipeline import run

run("pipeline.tvsm", "results/")
```

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

### Create Conda Environment

Source: https://tomviz.readthedocs.io/en/latest/ml_segmentation.html

Command to initialize the required conda environment for SAM 2 operations.

```bash
conda env create -f sam2-tomviz-cpu.yml
```

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

### tomviz.state.DataSource

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Creates a new data source from specified files.

```APIDOC
## tomviz.state.DataSource(fileNames)

### Description
Initializes a data source object using a list of file paths.

### Usage
```python
from tomviz import state
ds = state.DataSource(fileNames=['SMALL_PtCu_NanoParticle.tif'])
p = state.Pipeline(ds)
state.pipelines.append(p)
```
```

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

### Batch process datasets via command line

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Iterates through multiple files to process them using a saved pipeline state.

```bash
for f in dataset_*.emd; do
    tomviz-pipeline -s my_pipeline.tvh5 -d "$f" -o "output_${f}"
done
```

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

### Load a state file

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Use the state module to load an existing .tvsm state file into the application.

```python
>>> from tomviz import state
>>> state.load('test.tvsm')
```

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

### tomviz.utils.make_spreadsheet(column_names, table)

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/utils.html

Creates a vtkTable object from a 2D numpy array and a list of column names.

```APIDOC
## tomviz.utils.make_spreadsheet(column_names, table)

### Description
Converts a 2D numpy array into a vtkTable object, optionally including axes labels and log scale settings.

### Parameters
- **column_names** (list of strings) - Required - A list of names for the columns in the table.
- **table** (numpy.ndarray) - Required - A 2D numpy array containing the table data.
- **axes_labels** (list, optional) - Optional - A list of two strings for axes labels.
- **axes_log_scale** (list, optional) - Optional - A list of two booleans indicating log scale for axes.

### Returns
- **vtkTable** - A vtkTable object containing the formatted data.
```

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

### make_spreadsheet(column_names, table, axes_labels, axes_log_scale)

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/utils.html

Creates a vtkTable object that can be used within the Tomviz pipeline and saved to a JSON file.

```APIDOC
## make_spreadsheet(column_names, table, axes_labels, axes_log_scale)

### Description
Constructs a spreadsheet object (vtkTable) for use within Tomviz. If returned from an operator, it will appear in the pipeline.

### Parameters
- **column_names** (list[str]) - List of names for the table columns.
- **table** (np.ndarray) - The data to populate the table.
- **axes_labels** (tuple[str, str], optional) - Labels for the axes.
- **axes_log_scale** (tuple[bool, bool], optional) - Log scale settings for the axes.

### Returns
- **vtkTable** - The created spreadsheet object.
```

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

### Add a visualization

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Attach a Volume visualization module to a data source within a pipeline.

```python
>>> from tomviz import state
>>> ds = state.DataSource(
        fileNames=['SMALL_PtCu_NanoParticle.tif']
    )
>>> p = state.Pipeline(ds)
>>> state.pipelines.append(p)
>>> from tomviz.modules import Volume
>>> m = Volume()
>>> p.dataSource.modules.append(m)
```

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

### Implement a SourceNode

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Subclass tomviz.nodes.SourceNode and implement the produce method to generate data without inputs.

```python
import tomviz.nodes
import numpy as np

class MySphere(tomviz.nodes.SourceNode):
    def produce(self, radius=10.0, shape_x=100, shape_y=100, shape_z=100):
        ds = self.create_dataset()

        # Generate a sphere
        x, y, z = np.mgrid[:shape_x, :shape_y, :shape_z]
        center = np.array([shape_x, shape_y, shape_z]) / 2
        dist = np.sqrt((x - center[0])**2 + (y - center[1])**2 +
                       (z - center[2])**2)
        volume = (dist <= radius).astype(np.float32)

        ds.active_scalars = volume
        ds.spacing = (1.0, 1.0, 1.0)
        return {'output': ds}
```

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

### transform(dataset, threshold=127, radius_spacing=1, save_to_file=False, output_folder='')

Source: https://tomviz.readthedocs.io/en/latest/operators/utilities.html

Calculates the pore size distribution of a dataset.

```APIDOC
## transform(dataset, threshold=127, radius_spacing=1, save_to_file=False, output_folder='')

### Description
Calculates the pore size distribution of a dataset.

### Parameters
- **threshold** (int) - Optional - Default: 127, Minimum: 0
- **radius_spacing** (int) - Optional - Default: 1, Minimum: 1
- **save_to_file** (bool) - Optional - Default: False
- **output_folder** (directory) - Optional - Default: ''

### Results
- **pore_size_distribution** (table) - Pore Size Distribution
```

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

### tomviz.state.load

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Loads a Tomviz state file (.tvsm) into the application.

```APIDOC
## tomviz.state.load(filename)

### Description
Loads a state file into the current Tomviz session.

### Usage
```python
from tomviz import state
state.load('test.tvsm')
```
```

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

### Ptycho Source Configuration Defaults

Source: https://tomviz.readthedocs.io/en/latest/operators/utilities.html

Default values for Ptycho Source parameters including directory paths, file info, and dataset rotation settings.

```text
- default: ''
```

```text
- default: True
```

```text
- default: '[]'
```

```text
- default: '{}'
```

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

### PyStackRegImageAlignment.transform

Source: https://tomviz.readthedocs.io/en/latest/operators/alignment.html

Perform image alignment using PyStackReg.

```APIDOC
## PyStackRegImageAlignment.transform

### Description
Perform image alignment using PyStackReg.

### Parameters
- **dataset** (object) - The dataset to process.
- **transform_source** (enumeration) - The source of the transformation matrices. Options: 'generate', 'from_file'. Default: 'generate'.
- **transform_type** (enumeration) - Transformation type for the registration. Options: 'translation', 'rigid body', 'scaled rotation', 'affine', 'bilinear'. Default: 'translation'.
- **transform_file** (file) - Transformation file (npz format) to apply. Visible if transform_source is 'from_file'.
- **padding** (int) - Array padding along non-tilt axes before registration. Default: 10.
- **reference** (enumeration) - The reference slice to use. Options: 'previous', 'first', 'mean', 'slice_index'. Default: 'slice_index'.
- **ref_slice_index** (int) - Slice index to use for the reference. Default: 0.
- **apply_to_all_arrays** (bool) - Apply the registration transformation to all arrays within this data source. Default: True.
- **transforms_save_file** (save_file) - Save transformations file to apply to other datasets later.
```

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

### Handle Progress and Cancellation

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Use the progress and canceled attributes within a node to report status and handle user cancellation.

```python
class MyNode(tomviz.nodes.TransformNode):
    def transform(self, inputs, **params):
        self.progress.maximum = 100
        for i in range(100):
            if self.canceled:
                return None
            # Do work
            self.progress.value = i + 1
        return {'volume': inputs['volume']}
```

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

### Constant Dataset Configuration

Source: https://tomviz.readthedocs.io/en/latest/operators/utilities.html

Configuration parameters for generating a 3D dataset filled with a constant value.

```text
- default: `[100, 100, 100]`
- minimum: `[1, 1, 1]`
```

```text
- default: `0.0`
```

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

### view.save_screenshot(filename, palette=None, resolution=None)

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Saves the current view to an image file with optional palette and resolution settings.

```APIDOC
## view.save_screenshot(filename, palette=None, resolution=None)

### Description
Captures the active view and saves it to the specified filename.

### Parameters
- **filename** (str) - Required - The path to save the screenshot.
- **palette** (Palette) - Optional - The background palette to use (e.g., TransparentBackground, BlackBackground).
- **resolution** (tuple) - Optional - The resolution of the output image.
```

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

### OtsuMultipleThreshold

Source: https://tomviz.readthedocs.io/en/latest/operators/segmentation.html

Uses the Otsu multiple threshold algorithm to automatically determine thresholds.

```APIDOC
## OtsuMultipleThreshold.transform(dataset, number_of_thresholds=1, enable_valley_emphasis=False)

### Description
Use Otsu multiple threshold algorithm to automatically determine thresholds separating voxels into different classes based on image intensity.

### Parameters
- **dataset** (dataset) - Required - The input dataset.
- **number_of_thresholds** (int) - Optional - Number of thresholds (default: 1, minimum: 1).
- **enable_valley_emphasis** (bool) - Optional - Enable valley emphasis (default: False).

### Returns
- **label_map** (label_map) - Label Map
```

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

### tomviz.state.pipelines.execute

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Executes a pipeline associated with the current state.

```APIDOC
## pipeline.execute()

### Description
Triggers the execution of a pipeline object.

### Usage
```python
from tomviz import state
state.load('test.tvsm')
state.pipelines[0].execute()
```
```

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

### Random Particles Configuration Defaults

Source: https://tomviz.readthedocs.io/en/latest/operators/utilities.html

Default values for generating random 3D particles, including shape dimensions and sparsity settings.

```text
- default: [128, 128, 128]
- minimum: [1, 1, 1]
- maximum: [512, 512, 512]
```

```text
- default: 30.0
- minimum: 1.0
- maximum: 100.0
- step: 5.0
```

```text
- default: 60.0
- minimum: 1.0
- maximum: 100.0
- step: 5.0
```

```text
- default: 0.2
- minimum: 0.0
- maximum: 1.0
- step: 0.05
```

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

### Custom Transform Configuration JSON

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Define the display name, label, and description for a custom transform by creating a JSON file with the same base name as the Python script.

```json
{
  "name" : "Custom Thing",
  "label" : "Operate on data",
  "description" : "Apply my special operation to the data..."
}
```

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

### PyXRF Source Configuration Defaults

Source: https://tomviz.readthedocs.io/en/latest/operators/utilities.html

Default values for PyXRF Source parameters including command strings, directory paths, and processing flags.

```text
- default: 'pyxrf-utils'
```

```text
- default: ''
```

```text
- default: '[]'
```

```text
- default: False
```

```text
- default: 'sclr1_ch4'
```

```text
- default: True
```

```text
- default: '{}'
```

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

### Override inputs for single-source pipelines

Source: https://tomviz.readthedocs.io/en/latest/pipelines.html

Provide new input files or globs to replace the source defined in the state file.

```bash
# One file → one run.
tomviz-pipeline -s pipeline.tvsm -o results/ --input data.emd

# Glob → one run per matched file.
tomviz-pipeline -s pipeline.tvsm -o results/ --input 'data/*.emd'

# Explicit list (comma-separated, no spaces).
tomviz-pipeline -s pipeline.tvsm -o results/ --input a.emd,b.emd,c.emd
```

```python
from tomviz.pipeline import run

run("pipeline.tvsm", "results/", inputs="data.emd")
run("pipeline.tvsm", "results/", inputs="data/*.emd")
run("pipeline.tvsm", "results/", inputs=["a.emd", "b.emd", "c.emd"])
```

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

### Saving a screenshot with a transparent background

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Use the palette argument to specify a background style when saving a screenshot.

```python
>>> from tomviz import views
>>> view = views.active_view
>>> view.save_screenshot('screen.png',
                         palette=views.Palette.TransparentBackground)
```

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

### Accessing Tomviz views

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Retrieve the currently active view or a list of all available views from the state module.

```python
>>> from tomviz import state
>>> active = state.active_view
```

```python
>>> from tomviz import state
>>> views = state.views
```

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

### Create Spreadsheet Object

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/utils.html

Creates a vtkTable object for use within Tomviz, which can be displayed in the pipeline and exported to JSON.

```python
def make_spreadsheet(column_names: list[str], table: np.ndarray,
                     axes_labels: tuple[str, str] = None,
                     axes_log_scale: tuple[bool, bool] = None) -> 'vtkTable':
    """Make a spreadsheet object to use within Tomviz

    If returned from an operator, this will ultimately appear within the
    pipeline, and will be save-able to a JSON file.
```

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

### Execute a saved pipeline state

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Runs a saved pipeline file and writes the output to a specified file.

```bash
tomviz-pipeline -s <path_to_state_file> -o <path_to_write_output_emd>
```

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

### Configure External Subprocess Execution

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Specify a tomviz_pipeline_env path in the JSON metadata to run the transform in an isolated Python environment.

```json
{
  "name": "MyAITransform",
  "label": "AI Denoise",
  "tomviz_pipeline_env": "/path/to/conda/envs/ai_env"
}
```

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

### Create a simple Python transform

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Defines a basic transform function that operates on dataset scalars using NumPy.

```python
def transform(dataset):
    """Python transform that operates on the input array"""

    import numpy as np

    # Get the current volume as a numpy array.
    array = dataset.active_scalars

    # This is where you operate on your data, here we square root it.
    result = np.sqrt(array)

    # This is where the transformed data is set, it will display in tomviz.
    dataset.active_scalars = result

    # Optionally set the voxel sizes (in physical units)
    dataset.spacing = [5, 10, 7]
```

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

### Recon_TV_minimization.transform

Source: https://tomviz.readthedocs.io/en/latest/operators/reconstruction.html

Reconstruct a tilt series using TV Minimization.

```APIDOC
## Recon_TV_minimization.transform(dataset, Niter=10, Nupdates=0)

### Description
Reconstruct a tilt series using TV Minimization.

### Parameters
- **dataset** (dataset) - Required - The input tilt series dataset.
- **Niter** (int) - Optional - Number of iterations (default: 1, min: 0).
- **Nupdates** (int) - Optional - Number of updates (default: 0, min: 0, max: 100).
```

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

### Electron Beam Shape Parameters

Source: https://tomviz.readthedocs.io/en/latest/operators/utilities.html

Configuration parameters for generating a convergent electron beam in 3D for STEM imaging.

```text
- default: `300.0`
- minimum: `1.0`
- maximum: `1000000.0`
- step: `50.0`
```

```text
- default: `30.0`
- minimum: `0.0`
- maximum: `500.0`
- step: `0.5`
```

```text
- default: `256`
- minimum: `64`
- maximum: `2048`
```

```text
- default: `0.1`
- minimum: `0.0001`
- precision: 4 decimal places
- step: `0.1`
```

```text
- default: `512`
- minimum: `1`
- maximum: `2048`
```

```text
- default: `-50.0`
- minimum: `-1000000.0`
- maximum: `1000000.0`
- step: `5.0`
```

```text
- default: `100.0`
- minimum: `-1000000.0`
- maximum: `1000000.0`
- step: `5.0`
```

```text
- default: `0.2`
- minimum: `-1000000.0`
- maximum: `1000000.0`
- step: `0.1`
```

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

### Recon_tomopy.transform

Source: https://tomviz.readthedocs.io/en/latest/operators/reconstruction.html

Runs TomoPy reconstruction on a dataset using various algorithms.

```APIDOC
## Recon_tomopy.transform

### Description
Run TomoPy reconstruction on a dataset.

### Parameters
- **dataset** (dataset) - The input dataset to reconstruct.
- **algorithm** (enumeration) - Reconstruction algorithm to use (default: 'gridrec'). Options: 'gridrec', 'fbp', 'sirt', 'art', 'tv', 'mlem', 'ospml_hybrid'.
- **num_iter** (int) - Number of iterations for iterative methods (default: 5, range: 1-1000).
- **reg_par** (double) - Regularization parameter for TV (default: 0.1, range: 0.0001-10.0).
- **use_gpu** (bool) - Use GPU acceleration via CUDA (default: False).
```

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

### Dataset Methods

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/dataset.html

Methods for managing scalar arrays within the dataset.

```APIDOC
## Dataset Methods

### scalars(name: str | None = None) -> numpy.ndarray
- Retrieves a scalar array by name. If name is None, returns the active scalar array.
- Raises KeyError if the name does not exist.

### set_scalars(name: str, array: numpy.ndarray)
- Adds or updates a scalar array. If the name exists, it is overwritten.

### remove_scalars(name: str)
- Removes a scalar array by name.
- Raises KeyError if the name does not exist.
```

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

### tomviz.modules.Volume

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

Adds a Volume visualization module to a pipeline.

```APIDOC
## tomviz.modules.Volume()

### Description
Appends a Volume visualization module to the data source modules list.

### Usage
```python
from tomviz.modules import Volume
m = Volume()
p.dataSource.modules.append(m)
```
```

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

### Define an integer parameter

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Defines a numeric parameter with a minimum bound.

```json
{
  "name" : "iterations",
  "label" : "Number of Iterations",
  "type" : "int",
  "default" : 100,
  "minimum" : 0
}
```

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

### Define Custom Transform Metadata

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Use JSON to define transform names, labels, and user-configurable parameters.

```json
{
  "name": "Fancy Square Root",
  "label": "Classy Square Root",
  "description": "A configurable square root operator.",
  "parameters": [
    {
      "name": "number_of_chunks",
      "label": "Number of Chunks",
      "type": "int",
      "default": 10,
      "minimum": 1,
      "maximum": 1000
    }
  ]
}
```

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

### tomviz.nodes.SourceNode

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

A base class for nodes that produce output data without inputs. Users must implement the produce method.

```APIDOC
## tomviz.nodes.SourceNode

### Description
Base class for creating data sources. Implement the `produce` method to generate and return a dictionary of output datasets.

### Method
`produce(self, **kwargs)`

### Parameters
- **kwargs** (dict) - Keyword arguments passed from the JSON description file.

### Returns
- **dict** - A dictionary mapping output port names to Dataset objects.
```

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

### Apply a transformation to each scalar array

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/dataset.html

Demonstrates using apply_to_each_scalar_array within a TransformNode to apply a constant addition to all scalar arrays in a dataset.

```python
class AddConstant(tomviz.nodes.TransformNode):
    def transform(self, inputs, constant=0.0):
        ds = inputs["volume"]
        return {
            "volume": ds.apply_to_each_scalar_array(
                lambda a: a + constant)
        }
```

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

### Dataset Interface Methods

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/dataset.html

Methods and properties available on the Dataset object for data manipulation and metadata access.

```APIDOC
## Dataset.apply_to_each_scalar_array(fn)

### Description
Returns a new Dataset with each scalar array transformed by the provided function. The original dataset remains unmodified.

### Parameters
- **fn** (Callable[[np.ndarray], np.ndarray | None]) - Required - A function that takes a numpy array and returns a transformed array or None to exclude it.

### Returns
- **Dataset** - A new Dataset instance containing the transformed scalar arrays.

## Dataset.empty_copy()

### Description
Creates a new Dataset of the same type, preserving metadata (spacing, tilt angles, dimensions, file_name) but excluding scalar arrays.

### Returns
- **Dataset** - A new empty Dataset instance.

## Dataset Properties

- **scan_ids** (np.ndarray | None): Get or set the scan IDs for projections in a tilt series.
- **dark** (np.ndarray | None): Read-only access to dark field calibration data.
- **white** (np.ndarray | None): Read-only access to white field (flat field) calibration data.
- **file_name** (str | None): Read-only access to the original filename of the dataset.
- **metadata** (dict): Read-only access to the dictionary of arbitrary metadata associated with the dataset.
```

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

### Implement a TransformNode

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Subclass tomviz.nodes.TransformNode and implement the transform method to process input data.

```python
import tomviz.nodes

class AddConstant(tomviz.nodes.TransformNode):
    def transform(self, inputs, constant=0.0):
        ds = inputs['volume']
        ds.active_scalars = ds.active_scalars + constant
        return {'volume': ds}
```

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

### Define a multi-element integer parameter

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Defines an integer parameter that accepts an array of values.

```json
{
  "name" : "shift",
  "label" : "Shift",
  "type" : "int",
  "default" : [0, 0, 0]
}
```

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

### Dataset.apply_to_each_scalar_array(fn)

Source: https://tomviz.readthedocs.io/en/latest/api/index.html

Returns a new Dataset with each scalar array replaced by the result of the provided function. The original dataset remains unmodified.

```APIDOC
## Dataset.apply_to_each_scalar_array(fn)

### Description
Returns a new Dataset with each scalar array replaced by `fn(array)`. The original dataset is not modified.

### Parameters
- **fn** (Callable[[ndarray], ndarray | None]) - Required - A function that takes a numpy array and returns a new array or None to filter it out.

### Return Type
- **Dataset**
```

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

### BinaryOpen.transform

Source: https://tomviz.readthedocs.io/en/latest/operators/transformation.html

Perform morphological opening on segmented objects with a given label by a spherically symmetric structuring element of a given radius.

```APIDOC
## BinaryOpen.transform

### Description
Perform morphological opening on segmented objects with a given label by a spherically symmetric structuring element of a given radius.

### Parameters
- **dataset** (object) - The dataset to transform.
- **structuring_element_id** (enumeration) - Optional. 0: Box, 1: Ball, 2: Cross. Default: 0.
- **radius** (int) - Optional. Minimum: 1. Default: 1.
- **object_label** (int) - Optional. Default: 1.
- **background_label** (int) - Optional. Default: 0.
```

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

### Recon_WBP.transform

Source: https://tomviz.readthedocs.io/en/latest/operators/reconstruction.html

Reconstructs a tilt series using the Weighted Back Projection (WBP) method.

```APIDOC
## Recon_WBP.transform

### Description
Reconstruct a tilt series using Weighted Back Projection (WBP) method. The tilt series data should be aligned prior to reconstruction.

### Parameters
- **dataset** (dataset) - The input dataset.
- **Nrecon** (int) - Size of reconstruction (range: 1-4096).
- **filter** (enumeration) - Filter type (default: 1). Options: 0 (None), 1 (Ramp), 2 (Shepp-Logan), 3 (Cosine), 4 (Hamming), 5 (Hann).
- **interp** (enumeration) - Interpolation method (default: 0). Options: 0 (Linear), 1 (Nearest), 2 (Spline), 3 (Cubic).
- **Nupdates** (int) - Number of updates (default: 0, minimum: 0).
```

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

### Define a boolean parameter

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Defines a toggle parameter for the UI.

```json
{
  "name" : "enable_feature",
  "label" : "Enable Feature",
  "type" : "bool",
  "default" : false
}
```

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

### Shift3D.transform

Source: https://tomviz.readthedocs.io/en/latest/operators/alignment.html

Shift a dataset in 3D space.

```APIDOC
## Shift3D.transform

### Description
Shift a dataset.

### Parameters
- **dataset** (object) - The dataset to process.
- **SHIFT** (double) - Amount to shift by. Range: [-999, -999, -999] to [999, 999, 999]. Default: [0.0, 0.0, 0.0].
```

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

### tomviz.utils.pad_array(array, padding, tilt_axis)

Source: https://tomviz.readthedocs.io/en/latest/api/index.html

Adds padding to an array while ignoring the tilt axis.

```APIDOC
## tomviz.utils.pad_array(array, padding, tilt_axis)

### Description
Add padding to an array. Ignore the tilt axis.

### Parameters
- **array** (ndarray) - Required - The input array.
- **padding** (int) - Required - The amount of padding.
- **tilt_axis** (int) - Required - The axis to ignore.

### Return Type
- **ndarray**
```

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

### Recon_ART.transform

Source: https://tomviz.readthedocs.io/en/latest/operators/reconstruction.html

Reconstruct a tilt series using Algebraic Reconstruction Technique (ART) with a positivity constraint.

```APIDOC
## Recon_ART.transform(dataset, Niter=1, Nupdates=0, beta=1.0)

### Description
Reconstruct a tilt series using Algebraic Reconstruction Technique (ART) with a positivity constraint.

### Parameters
- **dataset** (dataset) - Required - The input tilt series dataset.
- **Niter** (int) - Optional - Number of iterations (default: 1, min: 1).
- **Nupdates** (int) - Optional - Number of updates (default: 0, min: 0, max: 100).
- **beta** (double) - Optional - Relaxation parameter (default: 1.0, min: 0.01, max: 2.0).
```

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

### Iterate Through All Dataset Channels

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Loop through all available scalar names to perform operations across multiple channels.

```python
def transform(dataset):
    import numpy as np

    channel_sum = None
    for name in dataset.scalars_names:
        channel = dataset.scalars(name)
        if channel_sum is None:
            channel_sum = channel
        else:
            channel_sum += channel

    dataset.active_scalars = channel_sum
```

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

### Define a double parameter

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

Defines a floating-point parameter with bounds and precision settings.

```json
{
  "name" : "rotation_angle",
  "label" : "Angle",
  "type" : "double",
  "default" : 90.0,
  "minimum" : -360.0,
  "maximum" : 360.0,
  "precision" : 1
}
```

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

### tomviz.Dataset

Source: https://tomviz.readthedocs.io/en/latest/operators_development.html

The Dataset object provides properties and methods for managing scalar arrays, metadata, and geometry.

```APIDOC
## tomviz.Dataset

### Properties
- **active_scalars** (ndarray) - Get/set the active scalar array.
- **active_name** (str) - Get/set the name of the active scalar.
- **num_scalars** (int) - Number of scalar arrays.
- **scalars_names** (list) - List of all scalar array names.
- **spacing** (tuple) - Voxel spacing (x, y, z).
- **tilt_angles** (ndarray) - NumPy array of tilt angles.
- **tilt_axis** (int/None) - Axis index for tilting.
- **scan_ids** (ndarray) - NumPy array of scan IDs.
- **dark/white** (data) - Dark/white field calibration data.
- **file_name** (str) - Original filename.
- **metadata** (dict) - Arbitrary metadata dictionary.

### Methods
- **scalars(name)** - Get a scalar array by name.
- **set_scalars(name, array)** - Add or update a scalar array.
- **remove_scalars(name)** - Remove a scalar array.
- **empty_copy()** - Create a new dataset with same geometry but no arrays.
```

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

### Defining screenshot palette options

Source: https://tomviz.readthedocs.io/en/latest/interactive.html

The Palette enum defines available background styles for saved screenshots.

```python
class Palette(Enum):
    Current = ""
    TransparentBackground = "TransparentBackground"
    BlackBackground = "BlackBackground"
    DefaultBackground = "DefaultBackground"
    GradientBackground = "GradientBackground"
    GrayBackground = "GrayBackground"
    PrintBackground = "PrintBackground"
    WhiteBackground = "WhiteBackground"
```

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

### Implement a CompletableOperator in Python

Source: https://tomviz.readthedocs.io/en/latest/_modules/tomviz/operators.html

Use the completed property to break out of iterative loops early while still returning the current dataset.

```python
import tomviz.operators

class MyCompletableOperator(tomviz.operators.CompletableOperator):
    def transform(self, dataset, ...):
        while True:
            if self.completed:
                # Break out of the iterative loop and return the current
                # results.
```

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

### SAM2Segment3D

Source: https://tomviz.readthedocs.io/en/latest/operators/segmentation.html

Segments a 3D volume with Meta’s Segment Anything Model 2 (SAM 2).

```APIDOC
## SAM2Segment3D.transform(dataset, seed=(-1, -1, -1), prompt_mode=0, invert_contrast=False, z_axis=0, propagate=0, model_size=2, device=0, checkpoint_dir='', trim_fraction=0.1, keep_seed_component=True)

### Description
Segment a 3D volume with Meta’s Segment Anything Model 2 (SAM 2). The Z-axis is treated as the video time axis.

### Parameters
- **dataset** (dataset) - Required - The input dataset.
- **seed** (int) - Optional - Seed point (default: [-1, -1, -1]).
- **prompt_mode** (enumeration) - Optional - Prompt mode (0: Point, 1: Auto Mask) (default: 0).
- **invert_contrast** (bool) - Optional - Invert contrast (default: False).
- **z_axis** (enumeration) - Optional - Z-axis (0: First, 1: Middle, 2: Last) (default: 0).
- **propagate** (enumeration) - Optional - Propagation direction (0: Both, 1: Forward, 2: Backward) (default: 0).
```
Tokens:
Raw