### Configure Mojmelo Setup Task in Pixi

Source: https://yetalit.github.io/Mojmelo/docs/Mojmelo/index

Defines a 'setup' task in 'pixi.toml' to install Mojmelo from source. This involves navigating to the Mojmelo directory and running the 'setup.mojo' script with various arguments.

```toml
[tasks]
setup = """
cd <path_to_mojmelo_location> &&
mojo ./setup.mojo &&
mojo ./setup.mojo 1 &&
mojo ./setup.mojo 2 &&
mojo ./setup.mojo 3 &&
mojo ./setup.mojo 4 &&
mojo ./setup.mojo 5 &&
mojo ./setup.mojo 6 &&
mojo ./setup.mojo 7 &&
mojo ./setup.mojo 8 &&
mojo ./setup.mojo 9"""
```

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

### Run Mojmelo Setup from Source using Pixi

Source: https://yetalit.github.io/Mojmelo/docs/Mojmelo/index

Executes the custom 'setup' task defined in 'pixi.toml' to install Mojmelo from its source code. This command triggers the execution of the 'setup.mojo' script.

```bash
pixi run setup
```

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

### Run Mojmelo Setup Script

Source: https://yetalit.github.io/Mojmelo/docs/Mojmelo/index

Executes the setup script for Mojmelo, which may include benchmarking and configuration steps. This script is typically found within the Pixi environment's conda test files.

```bash
bash ./.pixi/envs/default/etc/conda/test-files/mojmelo/0/tests/setup.sh
```

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

### Install Mojmelo using Pixi CLI

Source: https://yetalit.github.io/Mojmelo/docs/Mojmelo/index

This command adds the 'mojmelo' package to your project's dependencies using the Pixi CLI. Ensure the Modular community channel is configured in your 'pixi.toml'.

```bash
pixi add mojmelo
```

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

### Create Linspace Matrix

Source: https://yetalit.github.io/Mojmelo/docs/utils/Matrix/Matrix

Creates a matrix with values evenly spaced between a start and stop point. This static method requires the start, stop, number of points, and optionally an order parameter.

```Rust
@staticmethod
fn linspace(start: Float32, stop: Float32, num: Int, order: String = "c") -> Self
```

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

### Matrix Shape and Stride (`shape`, `stride`)

Source: https://yetalit.github.io/Mojmelo/docs/utils/mojmelo_matmul/matmul/Matrix

Methods to get the shape and stride information of the matrix.

```APIDOC
### `shape`

Returns the size of the matrix along a specified dimension.

```mojo
fn shape[dim: Int](self) -> Int
```
**Parameters:**
- **dim** (`Int`) - The dimension (0 for rows, 1 for columns) for which to get the size.
**Args:**
- **self** (`Self`) - The Matrix instance.
**Returns:**
- `Int` - The size of the matrix along the specified dimension.

### `stride`

Returns the stride of the matrix along a specified dimension.

```mojo
fn stride[dim: Int](self) -> Int
```
**Parameters:**
- **dim** (`Int`) - The dimension (0 for rows, 1 for columns) for which to get the stride.
**Args:**
- **self** (`Self`) - The Matrix instance.
**Returns:**
- `Int` - The stride of the matrix along the specified dimension.
```

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

### Get Data (`get_data`)

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/Cache

Retrieves data from the cache at a specific index.

```APIDOC
## GET /cache/get_data

### Description
Retrieves data from the cache at a specified index. This operation may modify the cache's LRU status.

### Method
GET

### Endpoint
/cache/get_data

### Parameters
#### Query Parameters
- **index** (`Int`) - Required - The index of the data to retrieve.
- **data** (`UnsafePointer[UnsafePointer[Float32, MutExternalOrigin], MutAnyOrigin]`) - Required - A pointer to where the retrieved data should be stored.
- **_len** (`Int`) - Required - The length of the data buffer to be filled.

### Request Example
```json
{
  "index": 5,
  "data": "<pointer_to_float32_array>",
  "_len": 100
}
```

### Response
#### Success Response (200)
- **Int** - The number of bytes successfully read.

#### Response Example
```json
{
  "bytes_read": 50
}
```
```

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

### HDBSCAN Initialization

Source: https://yetalit.github.io/Mojmelo/docs/HDBSCAN/HDBSCAN

Initializes the HDBSCAN struct with various parameters to control the clustering process. Default values are provided for most parameters, allowing for quick setup or fine-grained control over the clustering behavior.

```mojo
fn __init__(out self, min_samples: Int = 5, min_cluster_size: Int = 5, cluster_selection_method: String = "eom", alpha: Float32 = 1, cluster_selection_epsilon: Float32 = 0, cluster_selection_epsilon_max: Float32 = inf[DType.float32](), cluster_selection_persistence: Float32 = 0, max_cluster_size: Int = 0, allow_single_cluster: Bool = False, match_reference_implementation: Bool = False, search_deepness_coef: Int = 1)
__
```

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

### KDTreeBoruvka Node Building

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/KDTreeBoruvka/KDTreeBoruvka

The `build_node` method recursively builds a node in the KDTreeBoruvka, processing data within the specified range (`start` to `end`) for the given node index.

```mojo
fn build_node(mut self, node: Int, start: Int, end: Int)
__

**Args:**
  * **self** (`Self`)
  * **node** (`Int`)
  * **start** (`Int`)
  * **end** (`Int`)
```

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

### KDTreeBoruvka Split Dimension Choice

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/KDTreeBoruvka/KDTreeBoruvka

The `choose_split_dim` method determines the optimal dimension to split the data within a given range (`start` to `end`) using the provided index list.

```mojo
fn choose_split_dim(self, start: Int, end: Int, idx: List[Scalar[DType.index]]) -> Int
__

**Args:**
  * **self** (`Self`)
  * **start** (`Int`)
  * **end** (`Int`)
  * **idx** (`List`)


**Returns:**
`Int`
```

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

### Import Utility Modules in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/Mojmelo/index

Shows how to import utility classes and functions from the 'mojmelo.utils' package. This includes the 'Matrix' class and other functions imported with a wildcard.

```mojo
from mojmelo.utils.Matrix import Matrix
from mojmelo.utils.utils import *
```

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

### Matrix Initialization (`__init__`)

Source: https://yetalit.github.io/Mojmelo/docs/utils/mojmelo_matmul/matmul/Matrix

Details the different ways to initialize a Matrix object.

```APIDOC
## Methods: Matrix

### `__init__`

Initializes a new Matrix object.

#### Overload 1
```mojo
fn __init__(out self, shape: Tuple[Int, Int])
```
**Args:**
- **shape** (`Tuple`) - A tuple representing the desired shape (rows, columns) of the matrix.
- **self** (`Self`) - The instance of the Matrix to initialize.
**Returns:**
- `Self` - The initialized Matrix object.

#### Overload 2
```mojo
fn __init__(out self, data: UnsafePointer[Scalar[Type], MutAnyOrigin], var layout: Layout)
```
**Args:**
- **data** (`UnsafePointer`) - A pointer to the existing data for the matrix.
- **layout** (`Layout`) - The memory layout to be used for the matrix.
- **self** (`Self`) - The instance of the Matrix to initialize.
**Returns:**
- `Self` - The initialized Matrix object.

#### Overload 3
```mojo
fn __init__(out self, data: UnsafePointer[Scalar[Type], MutAnyOrigin], shape: Tuple[Int, Int])
```
**Args:**
- **data** (`UnsafePointer`) - A pointer to the existing data for the matrix.
- **shape** (`Tuple`) - A tuple representing the shape (rows, columns) of the matrix.
- **self** (`Self`) - The instance of the Matrix to initialize.
**Returns:**
- `Self` - The initialized Matrix object.
```

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

### KDTreeBoruvka Initialization

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/KDTreeBoruvka/KDTreeBoruvka

The `__init__` method for KDTreeBoruvka. It initializes the k-d tree with provided data, minimum samples, leaf size, and search depth coefficient.

```mojo
fn __init__(out self, data: Matrix, min_samples: Int, leaf_size: Int, search_deepness_coef: Int)
__

**Args:**
  * **data** (`Matrix`)
  * **min_samples** (`Int`)
  * **leaf_size** (`Int`)
  * **search_deepness_coef** (`Int`)
  * **self** (`Self`)


**Returns:**
`Self`
**Raises:**
```

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

### Get Cluster Tree Leaves

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/hdbscan_tree/get_cluster_tree_leaves

Retrieves the leaf nodes from a given cluster tree structure.

```APIDOC
## get_cluster_tree_leaves

### Description
Retrieves the leaf nodes from a given cluster tree structure.

### Method
POST

### Endpoint
/websites/yetalit_github_io_mojmelo/get_cluster_tree_leaves

### Parameters
#### Request Body
- **cluster_tree** (Dict) - Required - A dictionary representing the cluster tree. Keys are strings, and values are lists of scalar indices.

### Request Example
```json
{
  "cluster_tree": {
    "cluster1": [1, 2, 3],
    "cluster2": [4, 5]
  }
}
```

### Response
#### Success Response (200)
- **leaves** (List) - A list of scalar indices representing the leaves of the cluster tree.

#### Response Example
```json
{
  "leaves": [1, 2, 3, 4, 5]
}
```
```

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

### Mojo Solver Initialization and Status Checks

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/Solver

Provides methods for initializing the 'Solver' struct and checking the status of elements (lower bound, upper bound, free). These are fundamental operations for managing the state of the optimization problem.

```mojo
fn __init__(out self) -> Self
__

fn get_C(self, i: Int) -> Float64
__

fn update_alpha_status(self, i: Int)
__

fn is_upper_bound(self, i: Int) -> Bool
__

fn is_lower_bound(self, i: Int) -> Bool
__

fn is_free(self, i: Int) -> Bool
__
```

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

### Get Coefficient C in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/Solver_NU

The get_C method retrieves the coefficient C for a given index i from the Solver_NU struct.

```mojo
fn get_C(self, i: Int) -> Float64:
  # Implementation details for getting C
  pass

```

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

### SearchRecord Initialization

Source: https://yetalit.github.io/Mojmelo/docs/utils/KDTree/SearchRecord

Documentation for the `__init__` method of the SearchRecord struct, detailing its parameters and return type.

```APIDOC
## Method: SearchRecord.__init__

### Description
Initializes a new `SearchRecord` instance.

### Method
`__init__`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
*   **qv_in** (`Span`) - Input span for the query vector.
*   **tree_in** (`KDTree`) - The KD-tree to search within.
*   **result_in** (`KDTreeResultVector`) - The vector to store search results.
*   **self** (`Self`) - The instance of `SearchRecord` being initialized.

### Request Example
```python
# Example usage (conceptual, actual types may vary)
# search_record = SearchRecord(qv_in=span_data, tree_in=kd_tree, result_in=result_vector)
```

### Response
#### Success Response (200)
*   **Self** (`Self`) - The newly initialized `SearchRecord` instance.
```

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

### Get QD Vector in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/SVC_Q

Retrieves the QD vector from the SVC_Q struct. This method does not require mutable access.

```mojo
fn get_QD(self) -> UnsafePointer[Float64, MutExternalOrigin]:
  # Implementation details for getting QD vector
  pass

```

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

### Get Mojo Matrix Stride

Source: https://yetalit.github.io/Mojmelo/docs/utils/mojmelo_matmul/matmul/Matrix

Implements the `stride` method for the `Matrix` struct, returning the stride for a specified dimension index.

```mojo
fn stride[dim: Int](self) -> Int
__

**Parameters:**
  * **dim** (`Int`)

**Args:**
  * **self** (`Self`)

**Returns:**
`Int`
```

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

### Mojo KNN Constructor (__init__)

Source: https://yetalit.github.io/Mojmelo/docs/KNN/KNN

Initializes the KNN classifier. It can be initialized with default values for k and metric, or by providing a dictionary of parameters.

```mojo
fn __init__(out self, k: Int = 3, metric: String = "euc")
__

**Args:**
  * **k** (`Int`)
  * **metric** (`String`)
  * **self** (`Self`)


**Returns:**
`Self`
**Raises:**
```
fn __init__(out self, params: Dict[String, String])
__

**Args:**
  * **params** (`Dict`)
  * **self** (`Self`)


**Returns:**
`Self`
**Raises:**
```

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

### Get QD Data in SVR_Q Struct in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/SVR_Q

The `get_QD` method retrieves a pointer to the QD data from the SVR_Q struct.

```mojo
fn get_QD(self) -> UnsafePointer[Float64, MutExternalOrigin]
__
    # Args:
    #   self (Self)
    # Returns:
    #   UnsafePointer
__
```

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

### PCA Initialization and Fields

Source: https://yetalit.github.io/Mojmelo/docs/PCA/PCA

Details on how to initialize the PCA struct and a description of its available fields.

```APIDOC
## PCA Struct

### Description
Principal component analysis (PCA). Linear dimensionality reduction using Singular Value Decomposition of the data to project it to a lower dimensional space.

### Fields
- **n_components** (`Int`): Number of components to keep.
- **components** (`Matrix`)
- **components_T** (`Matrix`)
- **explained_variance** (`Matrix`): The amount of variance explained by each of the selected components.
- **explained_variance_ratio** (`Matrix`): Percentage of variance explained by each of the selected components.
- **mean** (`Matrix`)
- **whiten** (`Bool`): To transform data to have zero mean, unit variance, and no correlation between features.
- **whiten_** (`Matrix`)
- **lapack** (`Bool`): Use LAPACK to calculate svd.
```

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

### Get Mojo Matrix Shape

Source: https://yetalit.github.io/Mojmelo/docs/utils/mojmelo_matmul/matmul/Matrix

Implements the `shape` method for the `Matrix` struct, which returns the dimension size for a specified dimension index.

```mojo
fn shape[dim: Int](self) -> Int
__

**Parameters:**
  * **dim** (`Int`)

**Args:**
  * **self** (`Self`)

**Returns:**
`Int`
```

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

### GBDT Initialization in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/GBDT/GBDT

Provides two methods for initializing the GBDT model. The first allows direct parameter setting, while the second accepts parameters as a dictionary of strings. Both methods return the initialized GBDT instance.

```mojo
fn __init__(out self, criterion: String = "log", n_trees: Int = 10, min_samples_split: Int = 10, max_depth: Int = 3, learning_rate: Float32 = 0.10000000000000001, reg_lambda: Float32 = 1, reg_alpha: Float32 = 0, gamma: Float32 = 0, n_bins: Int = 0)
__
```

```mojo
fn __init__(out self, params: Dict[String, String])
__
```

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

### SIMD Load and Store (`load`, `store`)

Source: https://yetalit.github.io/Mojmelo/docs/utils/mojmelo_matmul/matmul/Matrix

Methods for loading and storing data using SIMD (Single Instruction, Multiple Data) operations.

```APIDOC
### `load`

Loads a SIMD vector of data from the matrix at the specified indices.

```mojo
fn load[width: Int, *, dim: Int](self, i: Int, j: Int) -> SIMD[Type, width]
```
**Parameters:**
- **width** (`Int`) - The width of the SIMD vector to load.
- **dim** (`Int`) - The dimension along which to load the SIMD vector.
**Args:**
- **self** (`Self`) - The Matrix instance.
- **i** (`Int`) - The row index.
- **j** (`Int`) - The column index.
**Returns:**
- `SIMD[Type, width]` - A SIMD vector containing the loaded data.

### `store`

Stores a SIMD vector of data into the matrix at the specified indices.

```mojo
fn store[width: Int, *, dim: Int](self, value: SIMD[Type, width], i: Int, j: Int)
```
**Parameters:**
- **width** (`Int`) - The width of the SIMD vector to store.
- **dim** (`Int`) - The dimension along which to store the SIMD vector.
**Args:**
- **self** (`Self`) - The mutable Matrix instance.
- **value** (`SIMD[Type, width]`) - The SIMD vector to store.
- **i** (`Int`) - The row index.
- **j** (`Int`) - The column index.
```

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

### delTree Function

Source: https://yetalit.github.io/Mojmelo/docs/DecisionTree/delTree

The delTree function recursively deletes a tree structure starting from the provided node. It takes an unsafe pointer to a Node as input.

```APIDOC
## delTree Function

### Description
Recursively deletes a tree structure starting from the provided node. It takes an unsafe pointer to a Node as input.

### Method
N/A (This is a function definition, not an HTTP endpoint)

### Endpoint
N/A

### Parameters
#### Path Parameters
N/A

#### Query Parameters
N/A

#### Request Body
N/A

### Request Example
N/A

### Response
#### Success Response (N/A)
N/A

#### Response Example
N/A
```

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

### PolyRegression Initialization

Source: https://yetalit.github.io/Mojmelo/docs/PolynomialRegression/PolyRegression

Provides two methods for initializing the PolyRegression model. The first allows direct setting of parameters like degree, learning rate, and regularization. The second accepts parameters as a dictionary of strings.

```mojo
fn __init__(out self, degree: Int = 2, learning_rate: Float32 = 0.01, n_iters: Int = 1000, reg_alpha: Float32 = 0, l1_ratio: Float32 = 0, tol: Float32 = 0, batch_size: Int = 0, random_state: Int = -1)

```

```mojo
fn __init__(out self, params: Dict[String, String])

```

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

### Get Data from Cache

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/Cache

Retrieves data from the 'Cache' at a specified 'index'. It populates the provided 'data' pointer and returns the length of the retrieved data.

```mojo
fn get_data(mut self, index: Int, data: UnsafePointer[UnsafePointer[Float32, MutExternalOrigin], MutAnyOrigin], var _len: Int) -> Int
__
```

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

### Get QD Vector in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/ONE_CLASS_Q

The 'get_QD' method for the 'ONE_CLASS_Q' struct in Mojo. It returns a pointer to a Float64 array representing the QD vector.

```mojo
fn get_QD(self) -> UnsafePointer[Float64, MutExternalOrigin]
__

**Args:**
  * **self** (`Self`)

**Returns:**
`UnsafePointer`
```

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

### Mojo CV __init__ Method

Source: https://yetalit.github.io/Mojmelo/docs/utils/utils/CV

Initializes the CV object with provided parameters. It takes a dictionary of string key-value pairs and returns the initialized object.

```mojo
fn __init__(out self: _Self, params: Dict[String, String])
__
```

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

### Get Max Value from KDTreeResultVector (Mojo)

Source: https://yetalit.github.io/Mojmelo/docs/utils/KDTree/KDTreeResultVector

The max_value method returns the maximum value currently stored in the KDTreeResultVector. It takes 'self' and returns a Float32.

```mojo
fn max_value(self) -> Float32

```

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

### KDTree Methods

Source: https://yetalit.github.io/Mojmelo/docs/utils/KDTree/KDTree

This section details the methods available for the KDTree struct, including initialization, tree building, and various nearest neighbor search operations.

```APIDOC
## KDTree Methods

### `__init__`

#### Description
Initializes a new KDTree.

#### Method
`__init__`

#### Parameters
- **X** (`Matrix`) - The input data matrix.
- **metric** (`String`) - The distance metric to use (e.g., "euc"). Defaults to "euc".
- **build** (`Bool`) - Whether to build the tree upon initialization. Defaults to `True`.
- **self** (`Self`) - The KDTree instance.

#### Returns
`Self` - The initialized KDTree instance.

### `__moveinit__`

#### Description
Move-initializes a KDTree from an existing one.

#### Method
`__moveinit__`

#### Parameters
- **existing** (`Self`) - The KDTree instance to move from.
- **self** (`Self`) - The KDTree instance to move to.

#### Returns
`Self` - The move-initialized KDTree instance.

### `__del__`

#### Description
Destroys the KDTree instance.

#### Method
`__del__`

#### Parameters
- **self** (`Self`) - The KDTree instance to destroy.

### `build_tree`

#### Description
Builds the KDTree from the existing data.

#### Method
`build_tree`

#### Parameters
- **self** (`Self`) - The KDTree instance.

### `build_tree_for_range`

#### Description
Builds a portion of the KDTree for a specified range.

#### Method
`build_tree_for_range`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **l** (`Int`) - The lower bound of the range.
- **u** (`Int`) - The upper bound of the range.
- **parent** (`UnsafePointer`) - The parent node.

#### Returns
`UnsafePointer` - A pointer to the newly created node.

### `spread_in_coordinate`

#### Description
Calculates the spread of points in a specific coordinate within a given range.

#### Method
`spread_in_coordinate`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **c** (`Int`) - The coordinate index.
- **l** (`Int`) - The lower bound of the range.
- **u** (`Int`) - The upper bound of the range.
- **interv** (`interval`) - The interval to update.

### `select_on_coordinate`

#### Description
Selects a pivot element on a specific coordinate within a given range.

#### Method
`select_on_coordinate`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **c** (`Int`) - The coordinate index.
- **k** (`Int`) - The desired rank.
- **l** (`Int`) - The lower bound of the range.
- **u** (`Int`) - The upper bound of the range.

### `select_on_coordinate_value`

#### Description
Selects elements based on a value threshold on a specific coordinate.

#### Method
`select_on_coordinate_value`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **c** (`Int`) - The coordinate index.
- **alpha** (`Float32`) - The threshold value.
- **l** (`Int`) - The lower bound of the range.
- **u** (`Int`) - The upper bound of the range.

#### Returns
`Int` - The number of elements selected.

### `n_nearest`

#### Description
Finds the N nearest neighbors to a query point.

#### Method
`n_nearest`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **qv** (`Span`) - The query point.
- **nn** (`Int`) - The number of nearest neighbors to find.
- **result** (`KDTreeResultVector`) - The vector to store the results.

### `n_nearest_around_point`

#### Description
Finds the N nearest neighbors around a specific point in the dataset.

#### Method
`n_nearest_around_point`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **idxin** (`Int`) - The index of the reference point.
- **correltime** (`Int`) - A time correlation parameter.
- **nn** (`Int`) - The number of nearest neighbors to find.
- **result** (`KDTreeResultVector`) - The vector to store the results.

### `r_nearest`

#### Description
Finds all neighbors within a given radius `r2` of a query point.

#### Method
`r_nearest`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **qv** (`Span`) - The query point.
- **r2** (`Float32`) - The squared radius.
- **result** (`KDTreeResultVector`) - The vector to store the results.

### `r_count`

#### Description
Counts the number of neighbors within a given radius `r2` of a query point.

#### Method
`r_count`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **qv** (`Span`) - The query point.
- **r2** (`Float32`) - The squared radius.

#### Returns
`Int` - The count of neighbors within the radius.

### `r_nearest_around_point`

#### Description
Finds all neighbors within a given radius `r2` around a specific point in the dataset.

#### Method
`r_nearest_around_point`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **idxin** (`Int`) - The index of the reference point.
- **correltime** (`Int`) - A time correlation parameter.
- **r2** (`Float32`) - The squared radius.
- **result** (`KDTreeResultVector`) - The vector to store the results.

### `r_count_around_point`

#### Description
Counts the number of neighbors within a given radius `r2` around a specific point in the dataset.

#### Method
`r_count_around_point`

#### Parameters
- **self** (`Self`) - The KDTree instance.
- **idxin** (`Int`) - The index of the reference point.
- **correltime** (`Int`) - A time correlation parameter.
- **r2** (`Float32`) - The squared radius.

#### Returns
`Int` - The count of neighbors within the radius.
```

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

### KDTreeBoruvka Initialization

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/KDTreeBoruvka/KDTreeBoruvka

Initializes a KDTreeBoruvka instance with the provided data and configuration parameters.

```APIDOC
## KDTreeBoruvka `__init__`

### Description
Initializes a KDTreeBoruvka instance.

### Method
`__init__`

### Endpoint
N/A (Constructor)

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
* **self** (`Self`) - The instance to initialize.
* **data** (`Matrix`) - The input data matrix.
* **min_samples** (`Int`) - The minimum number of samples required for a node.
* **leaf_size** (`Int`) - The maximum number of samples in a leaf node.
* **search_deepness_coef** (`Int`) - Coefficient for search deepness.

### Request Example
```python
# Example usage (conceptual)
# kd_tree = KDTreeBoruvka(data=my_matrix, min_samples=10, leaf_size=5, search_deepness_coef=3)
```

### Response
#### Success Response (200)
* **Self** (`Self`) - The initialized KDTreeBoruvka instance.

#### Response Example
```json
{
  "instance": "KDTreeBoruvka object"
}
```
```

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

### Access Column Slice with Offset in Mojo Matrix

Source: https://yetalit.github.io/Mojmelo/docs/utils/Matrix/Matrix

Retrieves a column slice with an offset, allowing access to sub-sections of a column based on starting index.

```mojo
fn __getitem__(self, offset: Bool, start_i: Int, column: Int) -> Self
__
```

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

### Cache Initialization (`__init__`)

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/Cache

Initializes a new Cache instance with specified length and size.

```APIDOC
## POST /cache/__init__

### Description
Initializes a new Cache instance with the specified length and size.

### Method
POST

### Endpoint
/cache/__init__

### Parameters
#### Query Parameters
- **l_** (`Int`) - Required - The initial length of the cache.
- **size_** (`Scalar[DType.uindex]`) - Required - The size of the cache.

### Request Example
```json
{
  "l_": 100,
  "size_": 1024
}
```

### Response
#### Success Response (200)
- **Self** (`Self`) - The newly created Cache instance.

#### Response Example
```json
{
  "__class__": "Cache",
  "l": 100,
  "size": 1024,
  "head": "<unsafe_pointer>",
  "lru_head": "<head_t_struct>"
}
```
```

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

### Access Row Slice with Offset in Mojo Matrix

Source: https://yetalit.github.io/Mojmelo/docs/utils/Matrix/Matrix

Retrieves a row slice with an offset, allowing access to sub-sections of a row based on starting index.

```mojo
fn __getitem__(self, row: Int, offset: Bool, start_i: Int) -> Self
__
```

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

### Initialize KMeans in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/KMeans/KMeans

Initializes the KMeans struct with specified parameters. Allows configuration of the number of clusters (K), initialization method (init), maximum iterations (max_iters), convergence criteria (converge), tolerance (tol), and random state for reproducibility.

```mojo
fn __init__(out self, K: Int = 5, init: String = "kmeans++", max_iters: Int = 100, converge: String = "centroid", tol: Float32 = 1.0E-4, random_state: Int = 42):
  self.K = K
  self.init = init
  self.max_iters = max_iters
  self.converge = converge
  self.tol = tol
  # Initialize other fields as needed, e.g., self.labels = []
  # Note: random_state is often used internally for initialization and not stored directly as a field in this example.
__
```

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

### Get Q Matrix Row in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/SVC_Q

Retrieves a specific row of the Q matrix from the SVC_Q struct. This method requires mutable access to the struct.

```mojo
fn get_Q(mut self, i: Int, _len: Int) -> UnsafePointer[Float32, MutExternalOrigin]:
  # Implementation details for getting Q matrix row
  pass

```

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

### Initialize RandomForest with Parameters in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/RandomForest/RandomForest

Provides two methods for initializing a RandomForest object. The first allows initialization with explicit parameters like n_trees, min_samples_split, max_depth, n_feats, criterion, and random_state. The second method initializes using a dictionary of string parameters.

```mojo
fn __init__(out self, n_trees: Int = 10, min_samples_split: Int = 2, max_depth: Int = 100, n_feats: Int = -1, criterion: String = "gini", random_state: Int = 42)

```

```mojo
fn __init__(out self, params: Dict[String, String])

```

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

### Get Length of KDTreeResultVector (Mojo)

Source: https://yetalit.github.io/Mojmelo/docs/utils/KDTree/KDTreeResultVector

The __len__ method returns the number of elements currently stored in the KDTreeResultVector. It takes 'self' and returns an Int representing the length.

```mojo
fn __len__(self) -> Int

```

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

### Mojo SVC Initialization

Source: https://yetalit.github.io/Mojmelo/docs/SVM/SVC

Provides two methods for initializing the SVC model. The first allows direct parameter setting, while the second accepts parameters as a dictionary of strings for flexible configuration.

```mojo
fn __init__(out self, C: Float64 = 0, nu: Float64 = 0, kernel: String = "rbf", degree: Int = 2, gamma: Float64 = -1, coef0: Float64 = 0, cache_size: Float64 = 200, tol: Float64 = 0.001, shrinking: Bool = True, probability: Bool = False, random_state: Int = -1)
__
```

```mojo
fn __init__(out self, params: Dict[String, String])
__
```

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

### PCA Methods

Source: https://yetalit.github.io/Mojmelo/docs/PCA/PCA

Documentation for the methods available in the PCA struct, including `__init__`, `fit`, `transform`, and `inverse_transform`.

```APIDOC
## PCA Methods

### `__init__`

#### Description
Initializes the PCA struct.

#### Method
`__init__`

#### Parameters
##### Path Parameters
None

##### Query Parameters
None

##### Request Body
- **n_components** (`Int`) - Required - Number of components to keep.
- **whiten** (`Bool`) - Optional - Whether to whiten the data (default: False).
- **lapack** (`Bool`) - Optional - Whether to use LAPACK for SVD (default: False).

### `fit`

#### Description
Fit the PCA model to the data.

#### Method
`fit`

#### Parameters
##### Path Parameters
None

##### Query Parameters
None

##### Request Body
- **X** (`Matrix`) - Required - The input data matrix.

### `transform`

#### Description
Apply dimensionality reduction to the input data.

#### Method
`transform`

#### Parameters
##### Path Parameters
None

##### Query Parameters
None

##### Request Body
- **X** (`Matrix`) - Required - The input data matrix to transform.

#### Response
##### Success Response (200)
- **Matrix** - The transformed data matrix.

### `inverse_transform`

#### Description
Transform the data back to its original space.

#### Method
`inverse_transform`

#### Parameters
##### Path Parameters
None

##### Query Parameters
None

##### Request Body
- **X_transformed** (`Matrix`) - Required - The transformed data matrix.

#### Response
##### Success Response (200)
- **Matrix** - The original data matrix.
```

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

### Get Length of SIMD Type (__len__)

Source: https://yetalit.github.io/Mojmelo/docs/utils/Matrix/Matrix

Returns the length (number of elements) of a SIMD type as an integer. This is a common method for determining the size of SIMD vectors.

```Rust
fn __len__(self) -> Int
__
```

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

### Initialize Mojo Matrix Struct

Source: https://yetalit.github.io/Mojmelo/docs/utils/mojmelo_matmul/matmul/Matrix

Provides multiple signatures for initializing the `Matrix` struct. These include initialization with shape, with data pointer and layout, and with data pointer and shape.

```mojo
fn __init__(out self, shape: Tuple[Int, Int])
__

**Args:**
  * **shape** (`Tuple`)
  * **self** (`Self`)

**Returns:**
`Self`
```

```mojo
fn __init__(out self, data: UnsafePointer[Scalar[Type], MutAnyOrigin], var layout: Layout)
__

**Args:**
  * **data** (`UnsafePointer`)
  * **layout** (`Layout`)
  * **self** (`Self`)

**Returns:**
`Self`
```

```mojo
fn __init__(out self, data: UnsafePointer[Scalar[Type], MutAnyOrigin], shape: Tuple[Int, Int])
__

**Args:**
  * **data** (`UnsafePointer`)
  * **shape** (`Tuple`)
  * **self** (`Self`)

**Returns:**
`Self`
```

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

### KMeans Initialization

Source: https://yetalit.github.io/Mojmelo/docs/KMeans/KMeans

Initializes the KMeans model with specified parameters for clustering.

```APIDOC
## POST /kmeans

### Description
Initializes the KMeans model with specified parameters for clustering.

### Method
POST

### Endpoint
/kmeans

### Parameters
#### Query Parameters
- **K** (Int) - Optional - The number of clusters to form as well as the number of centroids to generate. Defaults to 5.
- **init** (String) - Optional - Method for initialization -> 'kmeans++', 'random'. Defaults to 'kmeans++'.
- **max_iters** (Int) - Optional - Maximum number of iterations of the k-means algorithm for a single run. Defaults to 100.
- **converge** (String) - Optional - The converge method: Change in centroids <= tol -> 'centroid'; Change in inertia <= tol -> 'inertia'; Exact change in labels -> 'label'. Defaults to 'centroid'.
- **tol** (Float32) - Optional - Relative tolerance value. Defaults to 1.0E-4.
- **random_state** (Int) - Optional - Random seed for reproducibility. Defaults to 42.

### Request Body
This endpoint does not require a request body for initialization.

### Response
#### Success Response (200)
- **message** (String) - Confirmation message indicating successful initialization.

#### Response Example
```json
{
  "message": "KMeans model initialized successfully."
}
```
```

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

### Get Q Matrix Data in SVR_Q Struct in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/SVR_Q

The `get_Q` method retrieves a pointer to the Q matrix data for a given index `i` and length `_len`.

```mojo
fn get_Q(mut self, i: Int, _len: Int) -> UnsafePointer[Float32, MutExternalOrigin]
__
    # Args:
    #   self (Self)
    #   i (Int)
    #   _len (Int)
    # Returns:
    #   UnsafePointer
__
```

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

### Get Q Matrix Element in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/ONE_CLASS_Q

The 'get_Q' method for the 'ONE_CLASS_Q' struct in Mojo. It retrieves a pointer to a Float32 array representing a part of the Q matrix.

```mojo
fn get_Q(mut self, i: Int, _len: Int) -> UnsafePointer[Float32, MutExternalOrigin]
__

**Args:**
  * **self** (`Self`)
  * **i** (`Int`)
  * **_len** (`Int`)

**Returns:**
`UnsafePointer`
```

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

### Mojo Solver Methods for Working Set and Gradient Operations

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/Solver

Includes methods for swapping indices within the solver's state, reconstructing gradients, and selecting elements for the working set. These are crucial for iterative optimization algorithms.

```mojo
fn swap_index[QM: QMatrix](self, mut Q: QM, i: Int, j: Int)
__

fn reconstruct_gradient[QM: QMatrix](self, mut Q: QM)
__

fn select_working_set[QM: QMatrix](self, mut Q: QM, mut out_i: Int, mut out_j: Int) -> Int
__
```

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

### Mojo arange Function for Scalar Index Lists

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/hdbscan_tree/arange

Generates a list of scalar indices. It accepts integer start and stop values and returns a List of Scalar[DType.index].

```mojo
fn arange(start: Int, stop: Int) -> List[Scalar[DType.index]]
__

```

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

### Mojo LinearRegression Initialization

Source: https://yetalit.github.io/Mojmelo/docs/LinearRegression/LinearRegression

Provides two constructors for the `LinearRegression` struct. The first allows direct initialization of parameters, while the second accepts parameters as a dictionary of strings.

```mojo
fn __init__(out self, learning_rate: Float32 = 0.001, n_iters: Int = 1000, reg_alpha: Float32 = 0, l1_ratio: Float32 = -1, tol: Float32 = 0, batch_size: Int = 0, random_state: Int = -1)
__
```

```mojo
fn __init__(out self, params: Dict[String, String])
__
```

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

### Import Linear Regression Model in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/Mojmelo/index

Demonstrates how to import the LinearRegression class from the 'mojmelo.LinearRegression' module in a Mojo project.

```mojo
from mojmelo.LinearRegression import LinearRegression
```

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

### Get Components from TreeUnionFind in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/hdbscan_tree/TreeUnionFind

The `components` method returns a list of integers representing the disjoint components within the `TreeUnionFind` structure. It provides a snapshot of the current set partitioning.

```mojo
fn components(self) -> List[Int]

```

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

### KDTree Tree Building Methods in Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/KDTree/KDTree

Provides methods for building and managing the KDTree. `build_tree` constructs the entire tree, while `build_tree_for_range` constructs a portion of the tree within a specified range, returning a pointer to the root of the constructed subtree.

```mojo
fn build_tree(mut self)
__

fn build_tree_for_range(mut self, l: Int, u: Int, parent: UnsafePointer[KDTreeNode, MutAnyOrigin]) -> UnsafePointer[KDTreeNode, MutAnyOrigin]
__

```

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

### Get Element by Index in KDTreeResultVector (Mojo)

Source: https://yetalit.github.io/Mojmelo/docs/utils/KDTree/KDTreeResultVector

The __getitem__ method allows accessing an element at a specific index within the KDTreeResultVector. It takes 'self' and an 'index' (Int) and returns a KDTreeResult.

```mojo
fn __getitem__(self, index: Int) -> KDTreeResult

```

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

### KDTreeNode Methods

Source: https://yetalit.github.io/Mojmelo/docs/utils/KDTree/KDTreeNode

Provides documentation for the methods available on the KDTreeNode struct, including initialization, searching, and node processing.

```APIDOC
## Methods for KDTreeNode

### `__init__`

#### Description
Initializes a new KDTreeNode.

#### Method
`fn __init__(out self, dim: Int, metric: fn(Float32) -> Float32)`

#### Parameters
- **dim** (Int) - The dimension to use for splitting.
- **metric** (fn(Float32) -> Float32) - The metric function to be used.
- **self** (Self) - The instance of the KDTreeNode being initialized.

#### Returns
`Self` - The initialized KDTreeNode instance.

### `search`

#### Description
Performs a search operation within the KD-tree starting from this node.

#### Method
`fn search(self, mut sr: SearchRecord)`

#### Parameters
- **self** (Self) - The KDTreeNode instance.
- **sr** (SearchRecord) - The search record containing search criteria and results.

### `box_in_search_range`

#### Description
Checks if the bounding box of this node is within the current search range.

#### Method
`fn box_in_search_range(self, sr: SearchRecord) -> Bool`

#### Parameters
- **self** (Self) - The KDTreeNode instance.
- **sr** (SearchRecord) - The search record.

#### Returns
`Bool` - True if the bounding box is within the search range, False otherwise.

### `process_terminal_node`

#### Description
Processes a terminal node during a search operation.

#### Method
`fn process_terminal_node(self, mut sr: SearchRecord)`

#### Parameters
- **self** (Self) - The KDTreeNode instance.
- **sr** (SearchRecord) - The search record.

### `process_terminal_node_fixedball`

#### Description
Processes a terminal node specifically for a fixed-ball search.

#### Method
`fn process_terminal_node_fixedball(self, sr: SearchRecord)`

#### Parameters
- **self** (Self) - The KDTreeNode instance.
- **sr** (SearchRecord) - The search record.
```

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

### Mojo arange Function for Integer Lists

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/hdbscan_tree/arange

Generates a list of integers within a specified range. It takes a start and stop scalar value as input and returns a List of Int.

```mojo
fn arange(start: Scalar[DType.index], stop: Scalar[DType.index]) -> List[Int]
__

```

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

### LU Decomposition and Solving Linear Systems (Rust)

Source: https://yetalit.github.io/Mojmelo/docs/utils/Matrix/Matrix

Provides functions for LU decomposition, solving linear systems using LU factorization, and matrix inversion. These are static methods that operate on matrices and associated data.

```Rust
@staticmethod
fn lu_factor(mut A, piv: UnsafePointer[Int, MutAnyOrigin], N: Int)
__

```

```Rust
@staticmethod
fn lu_solve(A, piv: UnsafePointer[Int, MutAnyOrigin], b: Self, mut x: Self, N: Int, Mi: Int)
__

```

```Rust
@staticmethod
fn solve(var A, b: Self) -> Self
__

```

```Rust
fn inv(self) -> Self
__

```

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

### Mojo PCA Initialization

Source: https://yetalit.github.io/Mojmelo/docs/PCA/PCA

Constructor for the PCA struct. Initializes PCA with a specified number of components and optional parameters for whitening and LAPACK usage.

```mojo
fn __init__(out self, n_components: Int, whiten: Bool = False, lapack: Bool = False)

```

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

### Get Per Row for SIMD Types (get_per_row)

Source: https://yetalit.github.io/Mojmelo/docs/utils/Matrix/Matrix

Retrieves per-row data from a SIMD type, using another SIMD type as input for columns. It returns the SIMD type itself.

```Rust
fn get_per_row(self, columns: Self) -> Self
__
```

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

### bfs_from_cluster_tree Function

Source: https://yetalit.github.io/Mojmelo/docs/utils/hdbscan/hdbscan_tree/bfs_from_cluster_tree

This function implements a Breadth-First Search (BFS) algorithm on a given cluster tree structure. It starts the traversal from a specified root node and returns the order of visited nodes.

```APIDOC
## bfs_from_cluster_tree

### Description
Performs a Breadth-First Search (BFS) on a cluster tree starting from a given root node.

### Method
N/A (This is a function definition, not an HTTP endpoint)

### Endpoint
N/A

### Parameters
#### Path Parameters
N/A

#### Query Parameters
N/A

#### Request Body
N/A

### Request Example
N/A

### Response
#### Success Response (N/A)
- **List** (List[Scalar[DType.index]]) - The list of visited nodes in BFS order.

#### Response Example
N/A

### Args:
* **tree** (`Dict[String, List[Scalar[DType.index]]]`): The cluster tree represented as a dictionary where keys are node identifiers and values are lists of their neighbors.
* **bfs_root** (`Scalar[DType.index]`): The starting node for the BFS traversal.

### Returns:
`List[Scalar[DType.index]]`: A list containing the nodes visited during the BFS, in the order they were visited.

### Raises:
(Documentation for raised exceptions is not provided in the source text.)
```

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

### Mojo Layout Struct Initialization

Source: https://yetalit.github.io/Mojmelo/docs/utils/mojmelo_matmul/matmul/Layout

Provides two constructors for the 'Layout' struct: one taking both shape and strides, and another taking only shape. Both return an instance of 'Self'.

```mojo
fn __init__(shape: Tuple[Int, Int], strides: Tuple[Int, Int]) -> Self

```

```mojo
fn __init__(shape: Tuple[Int, Int]) -> Self

```

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

### Get SVM Type using Mojo

Source: https://yetalit.github.io/Mojmelo/docs/utils/libsvm/svm/svm_get_svm_type

The `svm_get_svm_type` function in Mojo takes an `svm_model` as input and returns an integer representing the SVM type. This function is useful for inspecting the properties of a trained SVM model.

```mojo
fn svm_get_svm_type(model: svm_model) -> Int
__
```