### Install and Run Tox

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Install tox with tox-uv and then run tox to verify the development installation. This ensures all project integrity checks pass.

```shell
uv tool install tox --with tox-uv
```

```shell
tox
```

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

### TOML Configuration Example

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/config_file.rst

Define basic client configuration parameters, including organization and credentials, in a TOML file. Ensure the TOML file is correctly formatted.

```toml
[default]
organization = "company"

[default.credentials]
username = "user@company.com"
password = "hunter12"
totp_enabled = true
```

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

### Install Pre-commit Hooks

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Install pre-commit and set up the git hooks to automatically run style checks before committing. This helps maintain code quality.

```bash
uv tool install pre-commit && pre-commit install
```

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

### Install PySimAI

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/installation.rst

Use this command to install or upgrade PySimAI. Ensure you have Python 3.9 or later.

```bash
pip install ansys-simai-core --upgrade
```

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

### Install PySimAI for Users

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Install the latest version of pip before installing PySimAI. This command installs the core package.

```bash
python -m pip install -U pip
```

```bash
python -m pip install ansys-simai-core
```

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

### Sequential Prediction and Postprocessing

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/best_practices.rst

This example demonstrates a sequential approach where data is accessed immediately after each computation, causing the client to wait for completion. This is less efficient for multiple requests.

```python
import ansys.simai.core as asc

simai_client = asc.from_config()
speeds = [5.9, 5.10, 5.11]

for geom in simai_client.geometries.list():
    for vx in speeds:
        # Run prediction
        pred = geom.run_prediction(Vx=vx)
        # Request global coefficients postprocessing
        # Because you are accessing the data, you must wait for the computation to finish
        coeffs = pred.post.global_coefficients().data
        # do something with the data
        print(coeffs)
```

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

### Efficient Prediction and Postprocessing

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/best_practices.rst

This example shows a more efficient method by initiating all predictions and postprocessing requests first, then waiting for all to complete using `simai_client.wait()`. Data is processed only after all computations are finished.

```python
import ansys.simai.core as asc

simai_client = asc.from_config()
speeds = [5.9, 5.10, 5.11]
predictions = []

for geom in simai_client.geometries.list():
    for vx in speeds:
        # Run prediction
        pred = geom.run_prediction(Vx=vx)
        # Request global coefficients postprocessing
        # Because you are not accessing the data, you are not blocked
        pred.post.global_coefficients()
        predictions.append(pred)

simai_client.wait()  # Wait for all objects requested locally to be complete
for pred in predictions:
    # do something with the data
    print(pred.post.global_coefficients().data)
```

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

### TOML Configuration with Profiles

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/config_file.rst

Configure multiple profiles within a single TOML file for different SimAIClient setups. Each profile can specify unique settings like organization and workspace.

```toml
[default]
organization = "company"
workspace = "my-usual-workspace"

[alternative]
organization = "company"
workspace = "some-other-workspace"
project = "red herring"
```

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

### Handle Detail Level Input Errors

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/automorphing/configure_automorphing.rst

These text examples show error messages that may occur if the 'detail_level' is too low or too high for the geometry and bounding box configuration.

```text
InputValueError: The size of box 'box_id' is too small for the chosen detail level.
The minimum detail level is 'min_detail_level'.
```

```text
InputValueError: The number of points to be deformed in the geometry ('nb_point') and
this detail level ('current_detail_level') might lead to Out of Memory. The maximum detail level is 'max_detail_level'.
```

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

### Process Exported CSV Data from ZIP Archive

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Read CSV data from a ZIP archive exported from global coefficients. This example shows how to use Python's zipfile and csv modules, and also demonstrates integration with pandas.

```python
import zipfile
import csv
from io import TextIOWrapper

data = selection.post.global_coefficients().export("csv.zip").in_memory()

with zipfile.ZipFile(data) as archive:
    csv_data = csv.reader(TextIOWrapper(archive.open("Global_Coeffs.csv")))

    # or with pandas:
    import pandas as pd

    df_geom = pd.read_csv(archive.open("Geometries.csv"))
```

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

### Build and Publish Package with uv

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Use uv commands to build the package for distribution or to publish it to a repository.

```bash
uv build
```

```bash
uv publish
```

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

### Build Documentation with Sphinx

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Build the project documentation using Sphinx and open the generated HTML file. This command is useful for previewing documentation changes locally.

```bash
uv run make -C doc/ html && open doc/html/index.html
```

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

### Build Documentation with Tox

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Use tox to build documentation and open the output. This is the recommended method for checking documentation integrity.

```bash
tox -e doc && open .tox/doc_out/index.html
```

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

### Create Training Data and Upload Folder

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Create a TrainingData instance and upload simulation data from a specified folder. This prepares data for project association.

```python
td = simai.training_data.create("my-first-data")
td.upload_folder("/path/to/folder/where/files/are/stored")
```

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

### Initialize SimAI Client

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Instantiate the SimAIClient to interact with the PySimAI service. You will be prompted for credentials upon execution.

```python
import ansys.simai.core as asc

simai_client = asc.SimAIClient()
```

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

### Create a Project

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Create a new project instance within SimAI. This project will be used to organize simulation data and configure AI models.

```python
project = simai.projects.create("my-first-project")
```

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

### Create SimAIClient from Default Config

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/config_file.rst

Instantiate the SimAIClient using the default configuration file path. The client automatically searches for the configuration in standard locations.

```python
import ansys.simai.core as asc

simai_client = asc.from_config()
```

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

### Create SimAIClient Instance

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/configuration.rst

Instantiate the SimAIClient by providing the organization name. Missing parameters will prompt user input.

```python
import ansys.simai.core as asc

simai_client = asc.SimAIClient(organization="my-company")
```

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

### Authenticate SimAIClient with Credentials

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/configuration.rst

Configure the SimAIClient with username and password for authentication. If credentials are not provided, the user will be prompted.

```python
import ansys.simai.core as asc

simai_client = asc.SimAIClient(
    organization="company",
    credentials={
        # neither of these are required, but if they are missing you will be
        # prompted to input them
        "username": "user@company.com",
        "password": "hunter12",
    },
)
```

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

### Create SimAIClient from Specific Config Path

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/config_file.rst

Instantiate the SimAIClient by providing a specific path to the configuration file. This allows you to manage configurations outside of default locations.

```python
simai_client = asc.from_config(path="/path/to/my/config")
```

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

### Load SimAIClient with a Specific Profile

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/config_file.rst

Instantiate the SimAIClient using a named profile from the configuration file. This is useful when you have multiple distinct configurations defined.

```python
simai_client = asc.from_config(profile="alternative")
```

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

### Build the AI Model

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Check if the project is trainable and then initiate the AI model build process using the configured ModelConfiguration. This requires the project to meet training requirements.

```python
if project.is_trainable():
     new_model = simai.models.build(mdl_conf)
```

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

### Configure the AI Model

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Create a ModelConfiguration instance, specifying project details, build preset, input/output definitions, global coefficients, and domain of analysis.

```python
mdl_conf = ModelConfiguration(
     project=project,                             # project of the model configuration
     build_preset="debug",                        # duration of the build
     build_on_top=False,                          # build on top of previous model
     input=model_input,                           # model input
     output=model_output,                         # model output
     global_coefficients=global_coefficients,     # Global Coefficients
     domain_of_analysis=doa                       # Domain of Analysis
)
```

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

### Run Lint and Test Tasks with Poe

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Alternatively, use uv and poethepoet to run development tasks like linting and testing. This provides a convenient way to manage common development workflows.

```shell
uv tool install poethepoet
```

```shell
uv run poe lint
```

```shell
uv run poe test
```

```shell
uv run poe doc
```

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

### Import Model Building Modules

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Import necessary classes for defining model configurations, including domain of analysis, inputs, and outputs.

```python
from ansys.simai.core.data.model_configuration import (
     DomainOfAnalysis,
     ModelConfiguration,
     ModelInput,
     ModelOutput,
)
```

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

### Download Binary Post-processing Results (VTU)

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Loop through a selection's post-processing results for volume VTU files and download each one individually.

```python
for vtu in selection.post.volume_vtu():
    vtu.data.download(f"/path/to/vtu_{vtu.id}")
```

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

### Creating and Running Predictions with Selections

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Demonstrates how to create a Selection by combining Geometries and Scalars, and then run predictions for all combinations within the selection.

```APIDOC
## Create and Run Predictions

### Description
This snippet shows how to instantiate a `Selection` object with a list of geometries and scalars, and then trigger the execution of predictions for all generated points.

### Method
```python
from ansys.simai.core.data.selections import Selection

geometries = simai.geometries.list()[:4]
scalars = [dict(Vx=vx) for vx in [12.2, 12.4, 12.6]]
selection = Selection(geometries, scalars)

# Run all predictions
selection.run_predictions()

all_predictions = selection.predictions
```

### Parameters
- `geometries` (list): A list of `Geometry` instances.
- `scalars` (list): A list of dictionaries, where each dictionary represents scalar values for a point.

### Returns
- `selection` (Selection): The created Selection object.
- `all_predictions` (list): A list of all generated `Prediction` instances after running `run_predictions()`.
```

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

### Create and Use a Selection

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/data_exploration.rst

Create a selection by combining geometries and scalars, then run predictions, compute global coefficients, and retrieve results.

```python
import ansys.simai.core as asc
from ansys.simai.core.data.selections import Selection

simai_client = asc.from_config()
geometries = simai_client.geometries.list()[:4]
scalars = [dict(Vx=vx) for vx in [12.2, 12.4, 12.6]]
selection = Selection(geometries, scalars)

# run all predictions
selection.run_predictions()

# compute all global coefficients
selection.post.global_coefficients()

# get all results
all_selection_coefficients = [
    global_coefficients.data
    for global_coefficients in selection.post.global_coefficients()
]
```

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

### Associate Training Data with Project

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Link the previously created training data to the specified project. This step is crucial before configuring the AI model.

```python
td.add_to_project(project)
```

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

### PostProcessingDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Manages the directory of available postprocessing objects. Use the `info` method to check availability.

```APIDOC
## PostProcessingDirectory

### Description
Provides access to and information about available postprocessing objects.

### Methods
- `info()`: Returns information about available postprocessing objects.
```

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

### LegacyOptimizationDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/optimizations.rst

Represents a directory for managing legacy optimization objects.

```APIDOC
## Class: LegacyOptimizationDirectory()

### Description
Provides methods for managing legacy optimization objects within a directory.

### Members
(Members are not explicitly listed here, refer to the source for details.)
```

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

### TrainingDataPartDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/training_data_parts.rst

Represents a directory containing training data parts. Provides methods for managing and accessing these parts.

```APIDOC
## TrainingDataPartDirectory

### Description
Represents a directory containing training data parts. Provides methods for managing and accessing these parts.

### Methods

* **__init__(self, path: str)**: Initializes the directory with a given path.
* **add_part(self, part: TrainingDataPart)**: Adds a TrainingDataPart to the directory.
* **get_part(self, part_name: str)**: Retrieves a TrainingDataPart by its name.
* **list_parts(self)**: Returns a list of all TrainingDataPart names in the directory.
* **remove_part(self, part_name: str)**: Removes a TrainingDataPart from the directory.
```

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

### ProjectDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/projects.rst

The ProjectDirectory class provides methods for managing and listing projects.

```APIDOC
## ProjectDirectory

### Description
Provides methods for managing and listing projects.

### Class
`ansys.simai.core.data.projects.ProjectDirectory`

### Members
(Details of members are not provided in the source text, but typically include methods for listing, creating, or deleting projects.)
```

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

### SimAIClient

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/client.rst

The SimAIClient class serves as the main entry point for interacting with the SimAI API. It encapsulates authentication, session management, and provides access to various SimAI functionalities.

```APIDOC
class SimAIClient
    """Client for the SimAI service.

    This class provides methods for interacting with the SimAI API, including
    authentication, project management, and simulation execution.
    """
    # Methods are documented here as they are defined in the source.
```

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

### OptimizationDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/optimizations.rst

Represents a directory for managing optimization objects.

```APIDOC
## Class: OptimizationDirectory()

### Description
Provides methods for managing optimization objects within a directory.

### Members
(Members are inherited and not explicitly listed here, refer to the source for details.)
```

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

### Clone PySimAI Repository

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Clone the PySimAI repository to begin development. Ensure Python is added to your system's Path if you are on Windows.

```bash
git clone https://github.com/ansys/pysimai
```

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

### TrainingDataDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/training_data.rst

Provides methods for managing training data directories, including uploading data.

```APIDOC
## TrainingDataDirectory()

### Description
Represents a directory for training data.

### Methods

#### upload_folder()

##### Description
Uploads data from a specified folder to be used as training data. This is the recommended method for standard workflows.

##### Parameters

* **folder_path** (str) - Required - The path to the folder containing the training data.

##### Returns

* A TrainingData object representing the uploaded data.
```

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

### Configure SimAIClient with Offline Token in TOML

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/configuration.rst

Configure SimAIClient using a TOML file, specifying organization, offline token, and disabling interactive mode for non-interactive authentication.

```toml
[default]
organization = "my-company"
offline_token = "your-offline-token-here"
interactive = false
```

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

### Accessing and Exporting Postprocessing Results

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Illustrates how to access postprocessing results through the `selection.post` namespace and export them in various formats like XLSX and CSV.

```APIDOC
## Postprocessing Operations

### Description
This section covers how to perform postprocessing operations on a selection, such as calculating global coefficients or surface evolution, and how to export these results.

### Methods

#### Global Coefficients

```python
# Get global coefficients
coeffs = selection.post.global_coefficients()

# Access the results
results_data = coeffs.data
```

#### Surface Evolution and Export

```python
# Calculate surface evolution and export to XLSX
selection.post.surface_evolution(axis="x", delta=13.5).export("xlsx").download(
    "/path/to/file.xlsx"
)

# Export global coefficients to a ZIP file containing CSVs
data = selection.post.global_coefficients().export("csv.zip").in_memory()

# Example of reading CSV data from the ZIP archive
import zipfile
import csv
from io import TextIOWrapper

with zipfile.ZipFile(data) as archive:
    csv_data = csv.reader(TextIOWrapper(archive.open("Global_Coeffs.csv")))
    # Process csv_data

    # Example using pandas
    import pandas as pd
    df_geom = pd.read_csv(archive.open("Geometries.csv"))
```

#### Downloading Binary Postprocessing Results

```python
# Download VTU files for volume postprocessing
for vtu in selection.post.volume_vtu():
    vtu.data.download(f"/path/to/vtu_{vtu.id}")
```

### Parameters
- `axis` (str): The axis for surface evolution calculation.
- `delta` (float): The delta value for surface evolution calculation.
- `format` (str): The export format (e.g., "xlsx", "csv.zip").
- `path` (str): The file path for downloading results.

### Returns
- `coeffs` (SelectionPostProcessingsMethods): An object representing postprocessing results.
- `data` (bytes): Binary data of the exported results.
- `vtu` (DownloadableResult): An object representing a downloadable VTU file.
```

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

### ModelConfiguration

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/models.rst

Defines the properties for building a SimAI model, including inputs, outputs, coefficients, and domain.

```APIDOC
class ModelConfiguration()

    (Details of ModelConfiguration properties and methods)
```

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

### IsTrainableInfo

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/projects.rst

Provides information about whether a project is trainable, excluding specific members.

```APIDOC
## IsTrainableInfo

### Description
Provides information about whether a project is trainable.

### Class
`ansys.simai.core.data.projects.IsTrainableInfo`

### Members
(Members are available, but `is_trainable` and `reason` are excluded from documentation.)
```

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

### Export Surface Evolution Data from Selection

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Export surface evolution data for a selection to an XLSX file and download it. For CSV exports, a ZIP file is generated.

```python
selection.post.surface_evolution(axis="x", delta=13.5).export("xlsx").download(
    "/path/to/file.xlsx"
)
```

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

### ContinuousLearningCapabilities

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/projects.rst

Represents the continuous learning capabilities of a project.

```APIDOC
## ContinuousLearningCapabilities

### Description
Represents the continuous learning capabilities of a project.

### Class
`ansys.simai.core.data.projects.ContinuousLearningCapabilities`

### Members
(Details of members are not provided in the source text.)
```

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

### TrainingCapabilities

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/projects.rst

Represents the training capabilities of a project.

```APIDOC
## TrainingCapabilities

### Description
Represents the training capabilities of a project.

### Class
`ansys.simai.core.data.projects.TrainingCapabilities`

### Members
(Details of members are not provided in the source text.)
```

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

### Build GeomAI Model

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/models.rst

Use the `build_model` method from `GeomAIProject` to create a new GeomAI model.

```APIDOC
## Build GeomAI Model

### Description
To build a GeomAI model, you should use the `build_model` method available on an instance of `GeomAIProject`.

### Method Signature
`GeomAIProject.build_model(config: GeomAIModelConfiguration)`

### Parameters
- **config** (`GeomAIModelConfiguration`): The configuration object specifying the parameters for the new GeomAI model.
```

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

### ModelDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/models.rst

Represents a directory for managing SimAI models. It provides methods for building and interacting with models.

```APIDOC
class ModelDirectory()

    Members:
        build(): Launches a build for a SimAI model using a ModelConfiguration.

    Exclude Members:
        get
```

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

### LegacyOptimization

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/optimizations.rst

Represents a single legacy optimization model.

```APIDOC
## Class: LegacyOptimization()

### Description
Represents a model for a legacy optimization process.

### Inherited Members
(Members are inherited from base classes and not explicitly listed here, refer to the source for details.)

### Members
(Members are not explicitly listed here, refer to the source for details.)
```

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

### Run Pytest with uv

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Use uv to run pytest for development. This command executes tests and provides verbose output.

```shell
uv run pytest -xlv
```

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

### TrainingDataPart

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/training_data_parts.rst

Represents a single file that is part of a larger TrainingData instance. Provides access to the file's content and metadata.

```APIDOC
## TrainingDataPart

### Description
Represents a single file that is part of a larger TrainingData instance. Provides access to the file's content and metadata.

### Methods

* **__init__(self, file_path: str, metadata: dict = None)**: Initializes the part with its file path and optional metadata.
* **get_content(self)**: Returns the content of the training data part.
* **get_metadata(self)**: Returns the metadata associated with the training data part.
* **set_metadata(self, key: str, value: any)**: Sets a metadata key-value pair.
* **save(self)**: Saves any changes made to the training data part.
```

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

### GeomAITrainingDataPart Model

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/training_data_parts.rst

This section details the GeomAITrainingDataPart class, representing a single file within a training data instance. Direct interaction is not recommended.

```APIDOC
## GeomAITrainingDataPart

### Description
Represents a singular file that is part of a GeomAITrainingData instance.

### Class
`GeomAITrainingDataPart()`

### Methods
(Members are inherited and detailed in the source documentation)
```

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

### GeomAIProjectDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/projects.rst

Represents a directory of GeomAI projects.

```APIDOC
## GeomAIProjectDirectory

### Description
Represents a directory of GeomAI projects. This class provides methods for managing and accessing projects within the GeomAI system.

### Class
`ansys.simai.core.data.geomai.projects.GeomAIProjectDirectory`

### Methods
(Members are documented in the source class)
```

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

### ModelConfiguration

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/model_configuration.rst

Defines the overall configuration for a model, including its inputs, outputs, and other settings.

```APIDOC
## ModelConfiguration()

### Description
Defines the overall configuration for a model, including its inputs, outputs, and other settings. This object is used to train a model.

### Parameters
This class does not explicitly list parameters in the source. Refer to the class definition for available attributes and methods.
```

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

### Project

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/projects.rst

The Project class represents a specific project, which is a selection of training data used to train a model. It inherits members from a base class.

```APIDOC
## Project

### Description
Represents a specific project, which is a selection of training data used to train a model.

### Class
`ansys.simai.core.data.projects.Project`

### Inherited Members
(The source indicates inherited members are available, but does not list them.)
```

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

### Generate Offline Token and Manage Consents

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/current_user.rst

Use this snippet to generate an offline token for the current user and manage their consents by listing or revoking them. Ensure the SimAI client is initialized from configuration.

```python
import ansys.simai.core as asc

simai_client = asc.from_config()

# Generate an offline token
token = simai_client.me.generate_offline_token()

# List all consents
consents = simai_client.me.consents.list()

# Revoke a specific consent
simai_client.me.consents.revoke("sdk")
```

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

### Tox Environments for Development Tasks

Source: https://github.com/ansys/pysimai/blob/main/README.rst

Utilize tox environments to automate common development tasks. These environments ensure isolated testing and quality checks.

```bash
tox -e style
```

```bash
tox -e py
```

```bash
tox -e py-coverage
```

```bash
tox -e doc
```

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

### GeomAITrainingDataDirectory Class

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/training_data.rst

Provides directory-level operations for GeomAITrainingData.

```APIDOC
## GeomAITrainingDataDirectory()

### Description
Initializes a GeomAITrainingDataDirectory object.

### Parameters
None

### Returns
A GeomAITrainingDataDirectory object.
```

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

### PostProcessing

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Base class for all postprocessing objects. Inherits common functionalities.

```APIDOC
## PostProcessing()

### Description
Represents a generic postprocessing object. This is a base class and typically used through its derived classes.

### Inherited Members
Includes members from `PostProcessing`.
```

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

### Model

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/models.rst

Represents a SimAI model. This class inherits properties and methods from its parent classes.

```APIDOC
class Model()

    Inherited Members:
        (Details of inherited members)
```

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

### PostProcessInput

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/model_configuration.rst

Defines the input parameters for post-processing steps.

```APIDOC
## PostProcessInput()

### Description
Defines the input parameters for post-processing steps.

### Parameters
This class does not explicitly list parameters in the source. Refer to the class definition for available attributes and methods.
```

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

### Configure Part Morphing Constraints

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/automorphing/configure_automorphing.rst

Use 'part_morphing' to define constraints for specific geometry parts. 'part_ids' identifies the parts, and 'continuity_constraint' (0-1) enforces interface smoothness.

```Python
from ansys.simai.core.data.optimizations import OptimizationPartMorphingSchema

part_morphing = OptimizationPartMorphingSchema(
    part_ids=[1, 2],
    continuity_constraint=0.8
)
```

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

### GeomAIWorkspaceDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/workspaces.rst

Represents a directory or collection of GeomAI workspaces.

```APIDOC
## GeomAIWorkspaceDirectory()

### Description
Provides access to a directory of GeomAI workspaces. This class likely allows listing, retrieving, or managing available workspaces.

### Method
(Not specified, assumed to be a class instantiation)

### Endpoint
(Not applicable for SDK methods)

### Parameters
(Parameters not explicitly detailed in the source)

### Request Example
```python
# Assuming GeomAIWorkspaceDirectory is importable
workspace_dir = GeomAIWorkspaceDirectory()
```

### Response
(Response details not explicitly detailed in the source)
```

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

### GeomAITrainingDataPart Directory Operations

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/training_data_parts.rst

This section covers the directory-related operations for GeomAITrainingDataPart. Note that direct interaction with parts is discouraged in favor of using higher-level methods.

```APIDOC
## GeomAITrainingDataPartDirectory

### Description
Provides directory-level operations for GeomAITrainingDataParts.

### Class
`GeomAITrainingDataPartDirectory()`

### Methods
(Members are inherited and detailed in the source documentation)
```

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

### ModelInput

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/model_configuration.rst

Defines the input parameters for the model.

```APIDOC
## ModelInput()

### Description
Defines the input parameters for the model.

### Parameters
This class does not explicitly list parameters in the source. Refer to the class definition for available attributes and methods.
```

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

### Define Model Inputs and Outputs

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Specify the input and output variables for the AI model. Inputs can include surface and scalar data, while outputs can be surface or volume data.

```python
model_input = ModelInput(surface=["wallShearStress"], scalars=["Vx"])

model_output = ModelOutput(surface=["alpha.water"], volume=["p", "p_rgh"])
```

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

### GeomAIPredictionConfiguration

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/predictions.rst

Configuration model for GeomAI predictions.

```APIDOC
## GeomAIPredictionConfiguration

### Description

This model defines the configuration parameters for GeomAI predictions.

### Fields

*   **model_name** (string) - Required - The name of the prediction model.
*   **version** (string) - Optional - The version of the prediction model.
*   **parameters** (dict) - Optional - Additional parameters for the prediction.
```

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

### GeometryDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geometries.rst

Represents a directory of geometries, providing methods to manage and access them.

```APIDOC
## GeometryDirectory()

### Description
Manages a collection of geometries within the SimAI platform.

### Class
GeometryDirectory
```

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

### Run Prediction on Geometry

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/predictions.rst

Use this snippet to initiate a prediction on a geometry object. The arguments for the prediction, such as 'Vx' for velocity, are model-dependent.

```python
# Run a prediction on a given geometry with the specified velocity.
velocity = 10.0
prediction = geometry.run_prediction(Vx=velocity)
```

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

### Set HTTPS Proxy in SimAIClient

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/proxy.rst

Manually set the HTTPS proxy when creating a SimAIClient instance. This overrides system proxy configurations.

```python
import ansys.simai.core as asc

simai_client = asc.SimAIClient(https_proxy="http://company_proxy:3128")
```

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

### Optimization

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/optimizations.rst

Represents a single optimization model.

```APIDOC
## Class: Optimization()

### Description
Represents a model for an optimization process.

### Inherited Members
(Members are inherited from base classes and not explicitly listed here, refer to the source for details.)

### Members
(Members are not explicitly listed here, refer to the source for details.)
```

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

### CustomVolumePointCloud

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Represents custom volume point cloud data as a postprocessing result.

```APIDOC
## CustomVolumePointCloud()

### Description
Represents custom volume data in a point cloud format, available as a postprocessing result.
```

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

### SurfaceVTP

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Represents surface data in VTP format as a postprocessing result.

```APIDOC
## SurfaceVTP()

### Description
Represents surface data in the Visualization Toolkit PolyData (VTP) format, accessible as a postprocessing result.
```

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

### SurfaceEvolution

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Represents surface evolution data as a postprocessing result.

```APIDOC
## SurfaceEvolution()

### Description
Represents data related to the evolution of surfaces, accessible as a postprocessing result.

### Inherited Members
Includes members from `PostProcessing`.
```

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

### Sweep Geometries for Local Optimization

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/data_exploration.rst

Use the geometry.sweep method to find geometries with metadata closest to a candidate geometry, useful for local optimization or gradient descent.

```python
geometry = simai.geometries.list()[0]
neighbour_geometries = geometry.sweep(swept_metadata=["height", "length"])

# with which a selection can be built:
selection = Selection(neighbour_geometries, [dict(Vx=13.4)])
```

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

### Define Domain of Analysis

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/pysimai_ug/building_a_model.rst

Configure the domain of analysis for the model, specifying relative ranges for length, width, and height.

```python
doa = DomainOfAnalysis(
     length=("relative_to_min", 15.321, 183.847),
     width=("relative_to_min", 1.034, 12.414),
     height=("relative_to_min", 2.046, 24.555),
)
```

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

### LegacyOptimizationResult

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/optimizations.rst

Represents the result of a legacy optimization.

```APIDOC
## Class: LegacyOptimizationResult()

### Description
Represents the outcome and data associated with a legacy optimization run.

### Members
(Members are not explicitly listed here, refer to the source for details.)
```

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

### Set Current Workspace

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/workspaces.rst

Configures the SimAIClient to use a specific workspace. This method is essential for directing client operations to the intended workspace context.

```APIDOC
## set_current_workspace

### Description
Sets the current workspace for the SimAIClient.

### Method
`SimAIClient.set_current_workspace()`

### Parameters
This method does not take any explicit parameters in its signature as presented, but it operates on the client's internal state to select a workspace.

### Usage
```python
client.set_current_workspace(workspace_name)
```

### Notes
Refer to the `WorkspaceDirectory` and `Workspace` classes for details on available workspaces and their properties.
```

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

### GeomAIModelConfiguration

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/models.rst

Defines the configuration parameters for a GeomAI model.

```APIDOC
## GeomAIModelConfiguration

### Description
Defines the configuration parameters for a GeomAI model. This Pydantic model specifies the structure and types of configuration settings required for model creation and operation.
```

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

### Configure TLS CA Bundle in TOML

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/proxy.rst

Specify the TLS CA bundle path in the configuration file to resolve SSL certificate verification errors when using a proxy. This is an alternative to setting the REQUESTS_CA_BUNDLE environment variable.

```toml
[default]
organization = "company"
tls_ca_bundle = "/home/username/Documents/my_company_proxy_ca_bundle.pem"
```

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

### Create a Selection in PySimAI

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Create a Selection by combining Geometry instances with Scalars. This forms a collection of Points for potential Prediction instances.

```python
from ansys.simai.core.data.selections import Selection

geometries = simai.geometries.list()[:4]
scalars = [dict(Vx=vx) for vx in [12.2, 12.4, 12.6]]
selection = Selection(geometries, scalars)
```

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

### TrainingData

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/training_data.rst

Represents a collection of training data parts for model training.

```APIDOC
## TrainingData()

### Description
A collection of TrainingDataPart instances representing a simulation for model training.

### Methods

(Inherited methods from base class are available)
```

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

### GeomAIClient

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/client.rst

The GeomAIClient class provides access to GeomAI functionalities. It is typically accessed through the nested GeomAI client inside SimAI.

```APIDOC
## GeomAIClient

### Description
The GeomAIClient class is the main entry point for GeomAI functionalities within the SimAI library. It is designed to be instantiated and used to perform various GeomAI-related operations.

### Instantiation
```python
from ansys.simai.core.data.geomai import GeomAIClient

# Instantiate the GeomAIClient
client = GeomAIClient()
```

### Methods
This class provides the following methods:

*   **`__init__()`**: Initializes the GeomAIClient.
*   **`__repr__()`**: Returns a string representation of the GeomAIClient instance.
*   **`__str__()`**: Returns a string representation of the GeomAIClient instance.

(Note: Specific method signatures and detailed descriptions for other members are not provided in the source text and are omitted here.)
```

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

### Run Prediction

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/predictions.rst

This method runs a SimAI-powered prediction on a given geometry. The arguments required, such as 'Vx' for velocity, are dependent on the specific model being used.

```APIDOC
## Run Prediction

### Description
Runs a SimAI-powered prediction on a given geometry with specified parameters.

### Method
`run_prediction`

### Parameters
* **Vx** (float) - Required - Velocity parameter for the prediction.

### Request Example
```python
velocity = 10.0
prediction = geometry.run_prediction(Vx=velocity)
```

### Response
* **Prediction** - An object representing the numerical prediction with geometry and scalars.
```

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

### Define Maximum Displacement for Bounding Boxes

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/automorphing/configure_automorphing.rst

When using the 'max_displacement' parameter, list values in the same order as the bounding boxes. Ensure units correspond to geometry coordinates.

```Python
bounding_boxes = [[0,1,0,2,0,4],[10,2,10,4,10,5]]
max_displacement = [0.002, 0.001]
```

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

### ModelOutput

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/model_configuration.rst

Defines the output parameters of the model.

```APIDOC
## ModelOutput()

### Description
Defines the output parameters of the model.

### Parameters
This class does not explicitly list parameters in the source. Refer to the class definition for available attributes and methods.
```

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

### DownloadableResult

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Helper class for managing downloadable results.

```APIDOC
## DownloadableResult()

### Description
A helper class that provides functionality for managing and downloading results.

### Members
(Members are documented within the source but not detailed here as they represent internal structure.)
```

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

### OptimizationPartMorphingSchema

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/optimizations.rst

Schema for defining part morphing within optimizations.

```APIDOC
## Class: OptimizationPartMorphingSchema()

### Description
Defines the schema for part morphing configurations in optimizations.

### Members
(Members are not explicitly listed here, refer to the source for details.)
```

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

### GeomAIModelDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/models.rst

Represents a directory of GeomAI models. Use this to manage collections of models.

```APIDOC
## GeomAIModelDirectory()

### Description
Represents a directory of GeomAI models. Use this to manage collections of models.

### Methods
- `__init__()`: Initializes a GeomAIModelDirectory.
- `add(model)`: Adds a GeomAIModel to the directory.
- `get(model_id)`: Retrieves a GeomAIModel by its ID.
- `list()`: Lists all GeomAIModels in the directory.
- `remove(model_id)`: Removes a GeomAIModel by its ID.
```

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

### List Consents

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/current_user.rst

Retrieves a list of all consents granted by the current user.

```APIDOC
## List Consents

### Description
Lists all consents granted by the current user.

### Method
`consents.list()`

### Endpoint
N/A (SDK method)

### Parameters
None

### Request Example
```python
consents = simai_client.me.consents.list()
```

### Response
- **consents** (list) - A list of consent objects.
```

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

### GeomAITrainingData Class

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/training_data.rst

Represents a collection of geometric training data parts. It supports .vtp and .stl file formats and requires watertight, manifold, and non-self-intersecting geometry.

```APIDOC
## GeomAITrainingData()

### Description
Initializes a GeomAITrainingData object, which is a collection of GeomAITrainingDataPart instances.

### Parameters
None

### Returns
A GeomAITrainingData object.
```

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

### Run Predictions for a Selection

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Execute all pending predictions for a given selection. Access all generated predictions after execution.

```python
# run all predictions
selection.run_predictions()

all_predictions = selection.predictions
```

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

### Manage Consents with SimAIClient

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/configuration.rst

List all active consents and revoke a specific consent using the SimAIClient. Revoking a consent invalidates associated offline tokens.

```python
# List all consents
consents = simai_client.me.consents.list()
for consent in consents:
    print(f"Client: {consent.client_id}, Created: {consent.created_date}")

# Revoke a specific consent (invalidates associated offline tokens)
simai_client.me.consents.revoke("sdk")
```

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

### ProcessGlobalCoefficientDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/global_coefficients_requests.rst

Represents a class for processing global coefficients within a directory. It likely handles multiple coefficient definitions or related files.

```APIDOC
## ProcessGlobalCoefficientDirectory

### Description
This class is designed to handle global coefficients within a directory context, potentially managing multiple coefficient definitions or batch processing.

### Usage
```python
# Example usage (assuming necessary imports and setup)
# coefficient_directory = ProcessGlobalCoefficientDirectory()
# ... methods for directory-based processing ...
```

### Methods
(Details of methods available via `:members:` directive in source)
```

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

### SurfaceVTPTDLocation

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Represents surface data in VTP format with T/D location information as a postprocessing result.

```APIDOC
## SurfaceVTPTDLocation()

### Description
Represents surface data in VTP format, including Time/Data location information, as a postprocessing result.
```

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

### GeomAIProject

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/projects.rst

Represents a single GeomAI project, which is a selection of training data.

```APIDOC
## GeomAIProject

### Description
Represents a GeomAI project, which is a selection of training data used to train a model. This class allows interaction with individual project data and configurations.

### Class
`ansys.simai.core.data.geomai.projects.GeomAIProject`

### Methods
(Members are documented in the source class, including inherited members)
```

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

### Slice

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Represents slice data as a postprocessing result.

```APIDOC
## Slice()

### Description
Represents data extracted from a slice, available as a postprocessing result.
```

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

### GeomAIPredictionDirectory

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geomai/predictions.rst

Provides methods for managing GeomAI prediction directories.

```APIDOC
## GeomAIPredictionDirectory()

### Description

This class provides methods for managing GeomAI prediction directories.

### Methods

*   `__init__()`: Initializes the GeomAIPredictionDirectory.
*   `list()`: Lists available prediction directories.
*   `get(id)`: Retrieves a specific prediction directory by its ID.
*   `create(name)`: Creates a new prediction directory.
*   `delete(id)`: Deletes a prediction directory by its ID.
```

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

### Generate Offline Token

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/configuration.rst

Generate a long-lived offline token using an authenticated SimAIClient. Store this token securely as it grants full account access.

```python
import ansys.simai.core as asc

simai_client = asc.SimAIClient(organization="my-company")

token = simai_client.me.generate_offline_token()
print(f"Store this token securely: {token}")
```

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

### Authenticate SimAIClient with Offline Token

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/configuration_guide/configuration.rst

Use an offline token for non-interactive authentication by providing it during client instantiation along with `interactive=False`. This is ideal for server-side scripts and automated workflows.

```python
import ansys.simai.core as asc

simai_client = asc.SimAIClient(
    organization="my-company",
    offline_token="your-offline-token-here",
    interactive=False,
)
```

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

### Access Global Coefficients from Selection Post-processing

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/selections.rst

Retrieve global coefficients for a selection. The results are stored in the 'data' attribute.

```python
coeffs = selection.post.global_coefficients()

coeffs.data  # is a list of results of each post-processings.
```

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

### VolumeVTU

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Represents volume data in VTU format as a postprocessing result.

```APIDOC
## VolumeVTU()

### Description
Represents volume data in the Visualization Toolkit Unstructured Grid (VTU) format, accessible as a postprocessing result.
```

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

### Generate Offline Token

Source: https://github.com/ansys/pysimai/blob/main/doc/source/user_guide/automorphing/configure_automorphing.rst

Generate an offline token to allow the server to authenticate on your behalf for server-side optimization. This token is valid for 30 days.

```Python
offline_token = simai_client.me.generate_offline_token()
```

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

### PredictionPostProcessings

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Namespace for nested prediction-related postprocessing operations.

```APIDOC
## PredictionPostProcessings()

### Description
Provides access to postprocessing operations specifically related to predictions. This acts as a nested namespace.

### Members
(Members are documented within the source but not detailed here as they represent internal structure.)
```

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

### GlobalCoefficients

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/post_processings.rst

Represents global coefficients as a postprocessing result.

```APIDOC
## GlobalCoefficients()

### Description
Represents global coefficients that can be accessed as a postprocessing result.

### Inherited Members
Includes members from `PostProcessing`.
```

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

### DomainOfAnalysis

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/model_configuration.rst

Defines the domain of analysis for the model.

```APIDOC
## DomainOfAnalysis()

### Description
Defines the domain of analysis for the model.

### Parameters
This class does not explicitly list parameters in the source. Refer to the class definition for available attributes and methods.
```

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

### Geometry

Source: https://github.com/ansys/pysimai/blob/main/doc/source/api_reference/geometries.rst

Represents a single 3D geometry model and its associated metadata.

```APIDOC
## Geometry()

### Description
Represents a 3D model and its associated metadata managed by the SimAI platform.

### Class
Geometry

### Inherited Members
This class inherits members from its parent class.
```