PySurvival

https://www.pysurvival.io/
Admin
PySurvival is an open source Python package for survival analysis modeling that predicts when events...
Tokens:46,336
Snippets:466
Trust Score:7.5
Update:5 months ago
Show doc for...
Tokens:
Raw
### Install GCC on MacOS using Homebrew

Source: https://www.pysurvival.io/installation.html

Use this command if you have never installed GCC before on MacOS.

```bash
brew install gcc

```

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

### Kaplan Meier Model Example

Source: https://www.pysurvival.io/models/kaplan_meier.html

Demonstrates how to initialize, fit, and visualize a Kaplan Meier survival model using sample data. Ensure matplotlib is installed for display.

```python
# Importing modules
import numpy as np
from matplotlib import pyplot as plt
from pysurvival.models.non_parametric import KaplanMeierModel
from pysurvival.utils.display import display_non_parametric
# %matplotlib inline #Uncomment when using Jupyter 

# Generating random times and event indicators 
T = np.round(np.abs(np.random.normal(10, 10, 1000)), 1)
E = np.random.binomial(1, 0.3, 1000)

# Initializing the KaplanMeierModel
km_model = KaplanMeierModel()

# Fitting the model 
km_model.fit(T, E, alpha=0.95)

# Displaying the survival function and confidence intervals
display_non_parametric(km_model)

```

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

### Example Usage of SmoothKaplanMeierModel

Source: https://www.pysurvival.io/models/smooth_kaplan_meier.html

Demonstrates initializing, fitting, and displaying the survival function of the SmoothKaplanMeierModel with sample data.

```python
# Importing modules
import numpy as np
from matplotlib import pyplot as plt
from pysurvival.models.non_parametric import SmoothKaplanMeierModel
from pysurvival.utils.display import display_non_parametric
# %matplotlib inline #Uncomment when using Jupyter 

# Generating random times and event indicators 
T = np.round(np.abs(np.random.normal(10, 10, 1000)), 1)
E = np.random.binomial(1, 0.3, 1000)

# Initializing the SmoothKaplanMeierModel
skm_model = SmoothKaplanMeierModel(bandwidth=1., kernel='Cosine')

# Fitting the model and display the survival function and confidence intervals
skm_model.fit(T, E, alpha=0.95)

# Displaying the survival function and confidence intervals
display_non_parametric(skm_model)

```

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

### Importing necessary libraries

Source: https://www.pysurvival.io/models/extra_survival_trees.html

Imports all required libraries for data simulation, model building, and evaluation. Ensure these are installed before running.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.survival_forest import ExtraSurvivalTreesModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
%pylab inline
```

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

### Install PySurvival via pip

Source: https://www.pysurvival.io/installation.html

The recommended method for installing PySurvival is using pip. This command will download and install the latest version from PyPI.

```bash
pip install pysurvival

```

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

### Build and Evaluate Survival Models with PySurvival

Source: https://www.pysurvival.io/index.html

This example demonstrates loading data, building CoxPH and LinearMultiTask models, fitting them to training data, and evaluating their performance using concordance index on test data. Ensure necessary modules are imported.

```python
# Loading the modules
from pysurvival.models.semi_parametric import CoxPHModel
from pysurvival.models.multi_task import LinearMultiTaskModel
from pysurvival.datasets import Dataset
from pysurvival.utils.metrics import concordance_index

# Loading and splitting a simple example into train/test sets
X_train, T_train, E_train, X_test, T_test, E_test = \
    Dataset('simple_example').load_train_test()

# Building a CoxPH model
coxph_model = CoxPHModel()
coxph_model.fit(X=X_train, T=T_train, E=E_train, init_method='he_uniform',
                l2_reg = 1e-4, lr = .4, tol = 1e-4)

# Building a MTLR model
mtlr = LinearMultiTaskModel()
mtlr.fit(X=X_train, T=T_train, E=E_train, init_method = 'glorot_uniform',
         optimizer ='adam', lr = 8e-4)

# Checking the model performance
c_index1 = concordance_index(model=coxph_model, X=X_test, T=T_test, E=E_test )
print("CoxPH model c-index = {:.2f}".format(c_index1))

c_index2 = concordance_index(model=mtlr, X=X_test, T=T_test, E=E_test )
print("MTLR model c-index = {:.2f}".format(c_index2))
```

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

### List GCC files on MacOS

Source: https://www.pysurvival.io/installation.html

This command helps in identifying the GCC installation path on MacOS.

```bash
brew ls gcc

```

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

### Install GCC on Ubuntu

Source: https://www.pysurvival.io/installation.html

Install GCC version 8 and its C++ compiler on Ubuntu. The version number can be changed to match your system's requirements.

```bash
sudo apt install gcc-8 g++-8

```

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

### Import Libraries for Survival Analysis Example

Source: https://www.pysurvival.io/models/random_survival_forest.html

Imports necessary libraries for data manipulation, modeling, and visualization in survival analysis. Includes pysurvival specific modules for simulations and models.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.survival_forest import RandomSurvivalForestModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
%pylab inline
```

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

### Import Necessary Libraries

Source: https://www.pysurvival.io/models/coxph.html

Imports essential libraries for data manipulation, modeling, and visualization. Ensure these are installed before running the code.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.semi_parametric import CoxPHModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
```

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

### Import necessary libraries for Kernel SVM example

Source: https://www.pysurvival.io/models/kernel_svm.html

Import required libraries including numpy, pandas, matplotlib, PySurvival models, simulation utilities, and scikit-learn for data splitting.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from pysurvival.models.svm import KernelSVMModel
from pysurvival.models.simulations import SimulationModel
from pysurvival.utils.metrics import concordance_index
from sklearn.model_selection import train_test_split
from scipy.stats.stats import pearsonr
# %pylab inline # to use in jupyter notebooks
```

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

### Import necessary packages for Linear MTLR example

Source: https://www.pysurvival.io/models/linear_mtlr.html

Imports required libraries for data manipulation, modeling, and evaluation, including numpy, pandas, matplotlib, scikit-learn, and pysurvival components.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.multi_task import LinearMultiTaskModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
#%matplotlib inline  # To use with Jupyter notebooks
```

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

### Upgrade GCC on MacOS using Homebrew

Source: https://www.pysurvival.io/installation.html

Use this command if you have already installed GCC and want to update to the latest version on MacOS.

```bash
brew upgrade gcc

```

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

### Importing Packages for Parametric Models

Source: https://www.pysurvival.io/models/parametric.html

Imports necessary libraries for using pysurvival's parametric models, including data manipulation, modeling, and plotting tools. Ensure all listed packages are installed.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.parametric import GompertzModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
%pylab inline

```

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

### Set C++ and C environment variables on Ubuntu

Source: https://www.pysurvival.io/installation.html

Assign the paths for the C++ and C compilers to the CC and CXX environment variables on Ubuntu. Ensure the version number matches your GCC installation.

```bash
export CXX=/usr/bin/g++-8
export CC=/usr/bin/gcc-8

```

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

### Importing necessary libraries for PySurvival and data simulation

Source: https://www.pysurvival.io/models/linear_svm.html

Imports required packages including numpy, pandas, matplotlib, PySurvival models and utilities, scikit-learn for data splitting, and scipy for statistical functions. This setup is essential for data generation and model training.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from pysurvival.models.svm import LinearSVMModel
from pysurvival.models.simulations import SimulationModel
from pysurvival.utils.metrics import concordance_index
from sklearn.model_selection import train_test_split
from scipy.stats.stats import pearsonr
# %pylab inline # to use in jupyter notebooks
```

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

### LinearSVMModel Instance and Fit Method

Source: https://www.pysurvival.io/models/linear_svm.html

Demonstrates how to create an instance of the LinearSVMModel and fit it to training data. The `fit` method trains the model using input samples, time-to-event data, and event indicators.

```APIDOC
## LinearSVMModel

### Description
Creates an instance of the Linear SVM model.

### Method Signature
`LinearSVMModel()`

## fit

### Description
Fit the estimator based on the given parameters.

### Method Signature
`fit(X, T, E, with_bias = True, init_method='glorot_normal', lr = 1e-2, max_iter = 100, l2_reg = 1e-4, tol = 1e-3, verbose = True)`

### Parameters
#### Path Parameters
- **X** (array-like) - Input samples; where the rows correspond to an individual sample and the columns represent the features (shape=[n_samples, n_features]).
- **T** (array-like) - Target values describing the time when the event of interest or censoring occurred.
- **E** (array-like) - Values that indicate if the event of interest occurred (E[i]=1 corresponds to an event, and E[i] = 0 means censoring, for all i).
- **with_bias** (bool) - Optional - Whether a bias should be added (default=True).
- **init_method** (str) - Optional - Initialization method to use. Options include 'glorot_uniform', 'he_uniform', 'uniform', 'glorot_normal', 'he_normal', 'normal', 'ones', 'zeros', 'orthogonal' (default = 'glorot_normal').
- **lr** (float) - Optional - Learning rate used in the optimization (default=1e-2).
- **max_iter** (int) - Optional - Maximum number of iterations in the Newton optimization (default=100).
- **l2_reg** (float) - Optional - L2 regularization parameter for the model coefficients (default=1e-4).
- **tol** (float) - Optional - Tolerance for stopping criteria (default=1e-3).
- **verbose** (bool) - Optional - Whether or not producing detailed logging about the modeling (default=True).

### Returns
- self : object
```

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

### Install GCC on CentOS 7

Source: https://www.pysurvival.io/installation.html

Install GCC and its C++ compiler on CentOS 7 using yum. The version number in 'devtoolset-8' can be adjusted based on the latest available GCC version.

```bash
sudo yum install centos-release-scl
sudo yum install devtoolset-8-gcc devtoolset-8-gcc-c++  -y

```

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

### Initialize SmoothKaplanMeierModel

Source: https://www.pysurvival.io/models/smooth_kaplan_meier.html

Instantiate the SmoothKaplanMeierModel with specified bandwidth and kernel type. Bandwidth controls smoothing degree, while kernel defines the smoothing function.

```python
SmoothKaplanMeierModel(bandwidth=0.1, kernel='normal')
```

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

### Import Libraries and Load Dataset

Source: https://www.pysurvival.io/tutorials/credit_risk.html

Imports necessary libraries and loads the German Credit dataset. Ensures the dataset is ready for analysis.

```python
# Importing modules
import pandas as pd
import numpy as np
from matplotlib import pyplot as plt
from pysurvival.datasets import Dataset
%pylab inline

# Reading the dataset
raw_dataset = Dataset('credit_risk').load()
print("The raw_dataset has the following shape: {}.".format(raw_dataset.shape))
raw_dataset.head(3)
```

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

### Importing necessary packages

Source: https://www.pysurvival.io/models/nonlinear_coxph.html

Imports all required libraries for data simulation, modeling, and evaluation.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.semi_parametric import NonLinearCoxPHModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
#%pylab inline
```

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

### KernelSVMModel Initialization

Source: https://www.pysurvival.io/models/kernel_svm.html

Initializes a KernelSVMModel instance with specified kernel and scaling parameters.

```APIDOC
## KernelSVMModel

### Description
Initializes a KernelSVMModel instance.

### Method
`__init__`

### Parameters
- **kernel** (str) - Optional - The type of kernel used to fit the model. Available kernels: Polynomial, Gaussian, Exponential, Tanh, Sigmoid, Rational Quadratic, Inverse Multiquadratic, Multiquadratic. (default="gaussian")
- **scale** (float) - Optional - Scale parameter of the kernel function. (default=1)
- **offset** (float) - Optional - Offset parameter of the kernel function. (default=0)
- **degree** (float) - Optional - Degree parameter of the polynomial/kernel function. (default=1)
- **auto_scaler** (bool) - Optional - Whether to automatically scale the features. (default=True)
```

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

### __init__

Source: https://www.pysurvival.io/models/parametric.html

Initializes the Parametric model. Allows setting the number of time bins and whether to automatically scale features.

```APIDOC
## __init__

### Description
Initializes the Parametric model with specified bins and auto-scaler settings.

### Parameters
- **bins** (int) - Optional - Number of subdivisions of the time axis. Defaults to 100.
- **auto_scaler** (boolean) - Optional - Determines whether a sklearn scaler should be automatically applied. Defaults to True.
```

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

### Importing necessary packages

Source: https://www.pysurvival.io/models/neural_mtlr.html

Imports all required libraries for data simulation, model building, and evaluation.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.multi_task import NeuralMultiTaskModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
%pylab inline
```

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

### Initialize LinearMultiTaskModel

Source: https://www.pysurvival.io/models/linear_mtlr.html

Initializes the LinearMultiTaskModel with specified number of bins and auto-scaler option. Use this to set up the model before fitting.

```python
LinearMultiTaskModel(bins=100, auto_scaler=True)
```

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

### Initialize SimulationModel

Source: https://www.pysurvival.io/models/simulations.html

Instantiate a SimulationModel with specified survival distribution, risk type, and parameters. Default values are used if not provided.

```python
SimulationModel( survival_distribution = 'exponential', risk_type = 'linear',
                 censored_parameter = 1., alpha = 1, beta = 1., bins = 100,
                 risk_parameter = 1.)
```

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

### SmoothKaplanMeierModel Initialization

Source: https://www.pysurvival.io/models/smooth_kaplan_meier.html

Initializes the SmoothKaplanMeierModel with specified bandwidth and kernel type.

```APIDOC
## SmoothKaplanMeierModel

### Description
Initializes the estimator with a specified bandwidth and kernel.

### Parameters
- **bandwidth** (double) - Optional (default=0.1) - Controls the degree of smoothing. Smaller values result in a curve closer to the original Kaplan Meier estimator but increase computation time. Larger values result in a smoother curve but may decrease accuracy.
- **kernel** (str) - Optional (default='normal') - Defines the type of kernel to use. Options include: 'Uniform', 'Epanechnikov', 'Normal', 'Biweight', 'Triweight', 'Cosine'.
```

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

### Set C++ and C environment variables on MacOS

Source: https://www.pysurvival.io/installation.html

Assign the paths for the C++ and C compilers to the CC and CXX environment variables on MacOS. Ensure to use the version number that matches your GCC installation.

```bash
export CXX=/usr/local/Cellar/gcc/8.2.0/bin/g++-8
export CC=/usr/local/Cellar/gcc/8.2.0/bin/gcc-8

```

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

### Prepare Training and Testing Datasets

Source: https://www.pysurvival.io/models/linear_mtlr.html

Splits the generated dataset into training and testing sets, and extracts features, time, and event columns for model input.

```python
features = sim.features

index_train, index_test = train_test_split( range(N), test_size = 0.2)
data_train = dataset.loc[index_train].reset_index( drop = True )
data_test  = dataset.loc[index_test].reset_index( drop = True )

X_train, X_test = data_train[features], data_test[features]
T_train, T_test = data_train['time'].values, data_test['time'].values
E_train, E_test = data_train['event'].values, data_test['event'].values
```

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

### Set C++ and C environment variables on CentOS 7

Source: https://www.pysurvival.io/installation.html

Assign the paths for the C++ and C compilers to the CC and CXX environment variables on CentOS 7. Adjust the path based on the installed GCC version.

```bash
export CXX=/opt/rh/devtoolset-8/root/usr/bin/x86_64-redhat-linux-g++
export CC=/opt/rh/devtoolset-8/root/usr/bin/x86_64-redhat-linux-gcc

```

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

### Initialize KernelSVMModel

Source: https://www.pysurvival.io/models/kernel_svm.html

Instantiate the KernelSVMModel with specified kernel and parameters. The `auto_scaler` parameter, when True, automatically scales the input features.

```python
KernelSVMModel(kernel = "gaussian", scale=1., offset=0., degree=1.,
    auto_scaler = True)
```

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

### LinearMultiTaskModel Initialization

Source: https://www.pysurvival.io/models/linear_mtlr.html

Initializes the LinearMultiTaskModel with specified parameters for time discretization and automatic scaling.

```APIDOC
## `__init__` - Initialization
```python
LinearMultiTaskModel(bins=100, auto_scaler=True)
```

### Parameters:
  * `bins`: **int** _(default=100)_ -- Number of subdivisions of the time axis 
  * `auto_scaler`: **bool** _(default=True)_ -- Determines whether a sklearn scaler should be automatically applied
```

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

### Creating and fitting the Linear SVM model

Source: https://www.pysurvival.io/models/linear_svm.html

Instantiates a LinearSVMModel and fits it to the training data (X_train, T_train, E_train). Customizes the fitting process with specific initialization methods, learning rate, and regularization parameters.

```python
svm_model = LinearSVMModel()
svm_model.fit(X_train, T_train, E_train, init_method='he_uniform',
    with_bias = True, lr = 0.5,  tol = 1e-3,  l2_reg = 1e-3)
```

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

### Initialize and Generate Simulation Data

Source: https://www.pysurvival.io/models/random_survival_forest.html

Initializes a SimulationModel with specified survival and risk parameters, then generates synthetic data for training and testing. Use this to create realistic survival data for model development.

```python
from pysurvival.models.simulation import SimulationModel

# Initializing the simulation model
sim = SimulationModel( survival_distribution = 'exponential',
                       risk_type = 'linear',
                       censored_parameter = 1,
                       alpha = 3)

# Generating N random samples 
N = 1000
dataset = sim.generate_data(num_samples = N, num_features=4)
```

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

### Initialize RandomSurvivalForestModel

Source: https://www.pysurvival.io/models/random_survival_forest.html

Initializes the RandomSurvivalForestModel with a specified number of trees. The default is 10 trees.

```python
RandomSurvivalForestModel(num_trees = 10)
```

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

### Initialize and Generate Simulation Data

Source: https://www.pysurvival.io/models/linear_mtlr.html

Initializes a SimulationModel with Weibull survival distribution and linear risk type, then generates a dataset with specified number of samples and features.

```python
sim = SimulationModel( survival_distribution = 'Weibull',
                       risk_type = 'linear',
                       censored_parameter = 10.0,
                       alpha = .01, beta = 3.0 )

N = 1000
dataset = sim.generate_data(num_samples = N, num_features = 3)
```

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

### Initialize and Generate Simulation Data

Source: https://www.pysurvival.io/models/parametric.html

Initializes a SimulationModel with specified parameters and generates a dataset. This is useful for creating synthetic data for testing survival models.

```python
sim = SimulationModel( survival_distribution = 'Gompertz',
                       risk_type = 'linear',
                       censored_parameter = 10.0,
                       alpha = .01, beta = 3.0 )

N = 1000
dataset = sim.generate_data(num_samples = N, num_features = 3)
```

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

### Instantiate and Fit Kernel SVM Model

Source: https://www.pysurvival.io/models/kernel_svm.html

Creates an instance of the KernelSVMModel with specified kernel and scale, then fits the model to the training data. Adjust 'lr', 'tol', and 'l2_reg' for optimal performance.

```python
svm_model = KernelSVMModel(kernel='Gaussian', scale=0.25)
svm_model.fit(X_train, T_train, E_train, init_method='orthogonal',
    with_bias = True, lr = 0.8,  tol = 1e-3,  l2_reg = 1e-4)
```

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

### Initialize ExtraSurvivalTreesModel

Source: https://www.pysurvival.io/models/extra_survival_trees.html

Initializes the ExtraSurvivalTreesModel with a specified number of trees. The default is 10 trees.

```python
ExtraSurvivalTreesModel(num_trees = 10)
```

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

### Import Libraries and Load Dataset

Source: https://www.pysurvival.io/tutorials/churn.html

Imports necessary libraries and loads the churn dataset using PySurvival. Displays the shape and first two rows of the raw dataset.

```python
# Importing modules
import pandas as pd
import numpy as np
from matplotlib import pyplot as plt
from pysurvival.datasets import Dataset
%pylab inline

# Reading the dataset
raw_dataset = Dataset('churn').load()
print("The raw_dataset has the following shape: {}.".format(raw_dataset.shape))
raw_dataset.head(2)
```

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

### Load and Inspect Maintenance Dataset

Source: https://www.pysurvival.io/tutorials/maintenance.html

Imports necessary libraries and loads the 'maintenance' dataset. Displays the shape and the first few rows of the raw data.

```python
# Importing modules
import pandas as pd
import numpy as np
from matplotlib import pyplot as plt
from pysurvival.datasets import Dataset
%pylab inline

# Reading the dataset
raw_dataset = Dataset('maintenance').load()
print("The raw_dataset has the following shape: {}.".format(raw_dataset.shape))
raw_dataset.head(3)
```

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

### Generating a simulated dataset

Source: https://www.pysurvival.io/models/neural_mtlr.html

Creates a dataset using a Log-Logistic parametric model with Gaussian risk. Ensure `num_samples` and `num_features` are set appropriately for your needs.

```python
# Initializing the simulation model
sim = SimulationModel( survival_distribution = 'Log-Logistic',
                       risk_type = 'gaussian',
                       censored_parameter = 10.1,
                       alpha = 0.1, beta=3.2 )

# Generating N random samples 
N = 1000
dataset = sim.generate_data(num_samples = N, num_features = 3)

# Showing a few data-points 
dataset.head(2)
```

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

### RandomSurvivalForestModel Initialization

Source: https://www.pysurvival.io/models/random_survival_forest.html

Initializes the RandomSurvivalForestModel with a specified number of trees.

```APIDOC
## RandomSurvivalForestModel

### Description
Initializes the estimator.

### Method Signature
```python
RandomSurvivalForestModel(num_trees = 10)
```

### Parameters
* **num_trees** (int) - Optional - number of trees that will be built in the forest. Defaults to 10.
```

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

### Generate Simulated Dataset

Source: https://www.pysurvival.io/models/coxph.html

Initializes a simulation model with specified survival distribution and risk type, then generates a dataset with a given number of samples and features. The 'head(2)' call displays the first two data points.

```python
# Initializing the simulation model
sim = SimulationModel( survival_distribution = 'log-logistic',
                       risk_type = 'linear',
                       censored_parameter = 10.1,
                       alpha = 0.1, beta=3.2 )

# Generating N random samples 
N = 1000
dataset = sim.generate_data(num_samples = N, num_features = 4)

# Showing a few data-points 
dataset.head(2)
```

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

### Fitting the ExtraSurvivalTreesModel

Source: https://www.pysurvival.io/models/extra_survival_trees.html

Initializes and fits the ExtraSurvivalTreesModel using the training data. Hyperparameters like `max_features`, `max_depth`, `min_node_size`, and `num_random_splits` can be tuned for optimal performance.

```python
# Building the model
xst = ExtraSurvivalTreesModel(num_trees=200)
xst.fit(X_train, T_train, E_train,
        max_features="sqrt", max_depth=5, min_node_size=20,
        num_random_splits = 1000)
```

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

### Compare Individual Repayment Speed Functions

Source: https://www.pysurvival.io/tutorials/credit_risk.html

Visualizes and compares the cumulative density functions (speed of repayment) for individuals across different risk groups. It highlights actual event times.

```python
# Initializing the figure
fig, ax = plt.subplots(figsize=(15, 5))

# Selecting a random individual that experienced an event from each group
groups = []
for i, (label, (color, indexes)) in enumerate(risk_groups.items()) :

    # Selecting the individuals that belong to this group
    if len(indexes) == 0 :
        continue
    X = X_test.values[indexes, :]
    T = T_test.values[indexes]
    E = E_test.values[indexes]

    # Randomly extracting an individual that experienced an event
    choices = np.argwhere((E==1.)).flatten()
    if len(choices) == 0 :
        continue
    k = np.random.choice( choices, 1)[0]

    # Saving the time of event
    t = T[k]

    # Computing the CDF for all times t
    cdf = 1. - neural_mtlr.predict_survival(X[k, :]).flatten()

    # Displaying the functions
    label_ = '{} risk'.format(label)
    plt.plot(neural_mtlr.times, cdf, color = color, label=label_, lw=2)
    groups.append(label)

    # Actual time
    plt.axvline(x=t, color=color, ls ='--')
    ax.annotate('T={:.1f}'.format(t), xy=(t, 0.5*(1.+0.2*i)),
        xytext=(t, 0.5*(1.+0.2*i)), fontsize=12)

# Show everything
groups_str = ', '.join(groups)
title = "Comparing cumulative density functions between {} risk grades"
title = title.format(groups_str)
plt.legend(fontsize=12)
plt.title(title, fontsize=15)
plt.xlim(0, 65)
plt.ylim(0, 1.05)
plt.show()
```

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

### SimulationModel Initialization

Source: https://www.pysurvival.io/models/simulations.html

Initializes a SimulationModel instance with specified survival distribution, risk type, and parameters.

```APIDOC
## SimulationModel

### Description
Initializes a SimulationModel instance.

### Method
__init__

### Parameters
- **survival_distribution** (string) - Optional (default = 'exponential') - Defines a known survival distribution. Available distributions: Exponential, Weibull, Gompertz, Log-Logistic, Log-Normal.
- **risk_type** (string) - Optional (default='linear') - Defines the type of risk function: Linear, Square, Gaussian.
- **censored_parameter** (double) - Optional (default = 1.) - Coefficient used to calculate the censored distribution (Normal distribution N(loc=censored_parameter, scale=5)).
- **alpha** (double) - Optional (default = 1.) - The scale parameter.
- **beta** (double) - Optional (default = 1.) - The shape parameter.
- **bins** (int) - Optional (default=100) - The number of bins of the time axis.
- **risk_parameter** (double) - Optional (default = 1.) - Scaling coefficient for the risk score. For linear: r(x)=exp(x⋅ω). For square: r(x)=exp(risk_parameter∗(x⋅ω)^2). For gaussian: r(x)=exp(e^-(x⋅ω)^2∗risk_parameter).

### Request Example
```python
from pysurvival.models.simulations import SimulationModel

model = SimulationModel(
    survival_distribution='weibull',
    risk_type='gaussian',
    alpha=2.0,
    beta=1.5,
    risk_parameter=0.5,
    bins=50
)
```
```

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

### ExtraSurvivalTreesModel Initialization

Source: https://www.pysurvival.io/models/extra_survival_trees.html

Initializes the ExtraSurvivalTreesModel with a specified number of trees.

```APIDOC
## ExtraSurvivalTreesModel(num_trees = 10)

### Description
Initializes the estimator.

### Parameters
* `num_trees` (int) - Optional - number of trees that will be built in the forest. Defaults to 10.
```

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

### Parametric Model Initialization

Source: https://www.pysurvival.io/models/parametric.html

Initializes a parametric model with specified bins and auto-scaler settings. The auto_scaler determines if a sklearn scaler is automatically applied.

```python
__init__(bins=100, auto_scaler=True)

```

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

### Generate simulated dataset

Source: https://www.pysurvival.io/models/kernel_svm.html

Initialize a SimulationModel with specified survival distribution and risk type, then generate N random samples with 4 features. The `dataset.head(2)` call displays the first two generated data points.

```python
# Initializing the simulation model
sim = SimulationModel( survival_distribution = 'Log-Logistic',
                       risk_type = 'square',
                       censored_parameter = 1.1,
                       alpha = 1.5, beta = 4)

# Generating N Random samples
N = 1000
dataset = sim.generate_data(num_samples = N, num_features = 4)

# Showing a few data-points
dataset.head(2)
```

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

### Prepare Training and Testing Data

Source: https://www.pysurvival.io/models/coxph.html

Splits the generated dataset into training and testing sets. It then separates features (X) from survival time (T) and event indicators (E) for both sets.

```python
# Defining the features
features = sim.features

# Building training and testing sets #
index_train, index_test = train_test_split( range(N), test_size = 0.2)
data_train = dataset.loc[index_train].reset_index( drop = True )
data_test  = dataset.loc[index_test].reset_index( drop = True )

# Creating the X, T and E input
X_train, X_test = data_train[features], data_test[features]
T_train, T_test = data_train['time'].values, data_test['time'].values
E_train, E_test = data_train['event'].values, data_test['event'].values
```

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

### Create and fit ConditionalSurvivalForestModel

Source: https://www.pysurvival.io/models/conditional_survival_forest.html

Initializes a ConditionalSurvivalForestModel with 200 trees and fits it to the training data using specified hyperparameters for tree construction.

```python
# Building the model
csf = ConditionalSurvivalForestModel(num_trees=200)
csf.fit(X_train, T_train, E_train,
        max_features="sqrt", max_depth=5, min_node_size=20,
        alpha = 0.05, minprop=0.1)
```

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

### Compare Model Predictions to Actual Values (Active Loans)

Source: https://www.pysurvival.io/miscellaneous/tips.html

This function visualizes the time series of actual and predicted numbers of active loans. It's similar to the repaid loans comparison but focuses on loans that are still active. Ensure you provide the correct model and test data.

```python
results = compare_to_actual(neural_mtlr, X_test, T_test, E_test,
                            is_at_risk = True,  figure_size=(16, 6),
                            metrics = ['rmse', 'mean', 'median'])
```

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

### Compare Overall Churn Predictions

Source: https://www.pysurvival.io/tutorials/churn.html

Compares the time series of actual and predicted churned customers using the `compare_to_actual` utility. Requires the trained model, test data (features, survival times, and event indicators), and specifies metrics for comparison.

```python
from pysurvival.utils.display import compare_to_actual
results = compare_to_actual(csf, X_test, T_test, E_test,
                            is_at_risk = False,  figure_size=(16, 6),
                            metrics = ['rmse', 'mean', 'median'])
```

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

### Import necessary libraries

Source: https://www.pysurvival.io/models/conditional_survival_forest.html

Imports all required libraries for data manipulation, modeling, and visualization.

```python
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from pysurvival.models.simulations import SimulationModel
from pysurvival.models.survival_forest import ConditionalSurvivalForestModel
from pysurvival.utils.metrics import concordance_index
from pysurvival.utils.display import integrated_brier_score
%pylab inline
```

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

### Initialize ConditionalSurvivalForestModel

Source: https://www.pysurvival.io/models/conditional_survival_forest.html

Initializes the ConditionalSurvivalForestModel with a specified number of trees. The default is 10 trees.

```python
ConditionalSurvivalForestModel(num_trees = 10)
```

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

### NonLinearCoxPH Initialization

Source: https://www.pysurvival.io/models/nonlinear_coxph.html

Initializes the NonLinearCoxPH model with a specified MLP structure and optional auto-scaler.

```APIDOC
## NonLinearCoxPH(structure, auto_scaler=True)

### Description
Initializes the NonLinearCoxPH model.

### Parameters
* `structure`: **list of dictionaries** -- Provides the structure of the MLP within the model.
  ex: `structure = [ {'activation': 'ReLU', 'num_units': 128}, {'activation': 'Tanh', 'num_units': 128}, ]`.
  Each dictionary corresponds to a fully connected hidden layer:
    * `num_units` is the number of hidden units in this layer
    * `activation` is the activation function that will be used. The list of all available activation functions can be found here.
  In case there are more than one dictionary, each hidden layer will be applied in the resulting MLP, using the order it is provided in the structure .
* `auto_scaler`: **boolean** _(default=True)_ -- Determines whether a sklearn scaler should be automatically applied
```

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

### Generate Survival Data with SimulationModel

Source: https://www.pysurvival.io/models/simulations.html

Initializes a SimulationModel with specified survival distribution, risk type, and parameters, then generates a synthetic dataset. Use this to create sample data for testing survival analysis models.

```python
import pandas as pd
from pysurvival.models.simulations import SimulationModel
%pylab inline

# Initializing the simulation model
sim = SimulationModel( survival_distribution = 'gompertz',
                       risk_type = 'linear',
                       censored_parameter = 5.0,
                       alpha = 0.01,
                       beta = 5., )

# Generating N Random samples
N = 1000
dataset = sim.generate_data(num_samples = N, num_features=5)

# Showing a few data-points
dataset.head(2)
```

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

### Configure Matplotlib backend on MacOS

Source: https://www.pysurvival.io/installation.html

This command creates or appends to the matplotlibrc file to set the backend to TkAgg, resolving potential Matplotlib issues on MacOS.

```bash
cd
echo "backend: TkAgg" >> ~/.matplotlib/matplotlibrc

```

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

### Generating dataset from a nonlinear Weibull model

Source: https://www.pysurvival.io/models/nonlinear_coxph.html

Initializes a SimulationModel with Weibull distribution and Gaussian risk type, then generates synthetic data. Use this to create realistic survival data for testing models.

```python
# Initializing the simulation model
sim = SimulationModel( survival_distribution = 'weibull',
                       risk_type = 'gaussian',
                       censored_parameter = 2.1,
                       alpha = 0.1, beta=3.2 )

# Generating N random samples 
N = 1000
dataset = sim.generate_data(num_samples = N, num_features=3)

# Showing a few data-points 
dataset.head(2)
```

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

### Fit Gompertz Parametric Model

Source: https://www.pysurvival.io/models/parametric.html

Initializes and fits a Gompertz parametric model to the training data. Various optimization parameters like learning rate, optimizer, regularization, and epochs can be specified.

```python
from pysurvival.models.parametric import GompertzModel

gomp_model = GompertzModel()
gomp_model.fit(X_train, T_train, E_train, lr=1e-2, init_method='zeros',
    optimizer ='adam', l2_reg = 1e-3, num_epochs=2000)
```

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

### Displaying baseline simulations

Source: https://www.pysurvival.io/models/nonlinear_coxph.html

Visualizes the base survival function of the simulation model. This helps understand the underlying survival characteristics of the generated data.

```python
from pysurvival.utils.display import display_baseline_simulations
display_baseline_simulations(sim, figure_size=(10, 5))
```

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

### Fit KernelSVMModel

Source: https://www.pysurvival.io/models/kernel_svm.html

Fit the Kernel SVM model to the provided data. Ensure `X`, `T`, and `E` are correctly formatted. The `init_method` parameter controls the weight initialization strategy.

```python
fit(X, T, E, with_bias = True, init_method='glorot_normal', lr = 1e-2,
    max_iter = 100, l2_reg = 1e-4, tol = 1e-3, verbose = True)
```

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

### Compare Model Predictions to Actual Values (Repaid Loans)

Source: https://www.pysurvival.io/miscellaneous/tips.html

Use this function to visualize the time series of actual and predicted numbers of fully repaid loans. It requires the trained model, test data features, survival times, and event indicators. Metrics like RMSE, mean, and median can be calculated.

```python
from pysurvival.utils.display import compare_to_actual
results = compare_to_actual(neural_mtlr, X_test, T_test, E_test,
                            is_at_risk = False,  figure_size=(16, 6),
                            metrics = ['rmse', 'mean', 'median'])
```

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

### Import Libraries and Load Dataset

Source: https://www.pysurvival.io/tutorials/employee_retention.html

Imports necessary libraries (pandas, numpy, matplotlib) and the PySurvival Dataset module. Loads the 'employee_attrition' dataset and prints its shape. Use this to begin data loading and initial inspection.

```python
# Importing modules
import pandas as pd
import numpy as np
from matplotlib import pyplot as plt
from pysurvival.datasets import Dataset
%pylab inline

# Reading the dataset
raw_dataset = Dataset('employee_attrition').load()
print("The raw_dataset has the following shape: {}.".format(raw_dataset.shape))
raw_dataset.head(3)
```

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

### ConditionalSurvivalForestModel Initialization

Source: https://www.pysurvival.io/models/conditional_survival_forest.html

Initializes the Conditional Survival Forest model with a specified number of trees.

```APIDOC
## ConditionalSurvivalForestModel

### Description
Initializes the estimator.

### Parameters
* `num_trees` (int, default=10) - The number of trees to build in the forest.
```

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

### Display Baseline Simulation Data and Survival Function

Source: https://www.pysurvival.io/models/random_survival_forest.html

Displays the first few data points of the generated dataset and visualizes the base survival function of the simulation model. This helps in understanding the underlying data distribution and model assumptions.

```python
from pysurvival.utils.display import display_baseline_simulations

# Showing a few data-points 
print(dataset.head(2))

# Displaying the Base Survival function of the Simulation model 
display_baseline_simulations(sim, figure_size=(20, 6))
```

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

### RandomSurvivalForestModel Instantiation

Source: https://www.pysurvival.io/models/random_survival_forest.html

To create an instance of the Random Survival Forest model, use the `RandomSurvivalForestModel` class from `pysurvival.models.survival_forest`.

```APIDOC
## Instantiate RandomSurvivalForestModel

### Description
Creates an instance of the Random Survival Forest model.

### Method
```python
RandomSurvivalForestModel()
```

### Parameters
This method does not take any parameters for instantiation.
```

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

### Fit Cox PH Model

Source: https://www.pysurvival.io/models/coxph.html

Initializes a CoxPHModel and fits it to the training data. Parameters like learning rate (lr), L2 regularization (l2_reg), and initialization method (init_method) can be specified during fitting.

```python
# Building the model
coxph = CoxPHModel()
coxph.fit(X_train, T_train, E_train, lr=0.5, l2_reg=1e-2, init_method='zeros')
```

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

### Prepare Data for Kernel SVM

Source: https://www.pysurvival.io/models/kernel_svm.html

Splits the dataset into training and testing sets and extracts features, time, and event data. Ensure 'features', 'time', and 'event' columns exist in your dataset.

```python
index_train, index_test = train_test_split( range(N), test_size = 0.2)
data_train = dataset.loc[index_train].reset_index( drop = True )
data_test  = dataset.loc[index_test].reset_index( drop = True )

X_train, X_test = data_train[features], data_test[features]
T_train, T_test = data_train['time'].values, data_test['time'].values
E_train, E_test = data_train['event'].values, data_test['event'].values
```
Tokens:
Raw