feyn.reference

This module contains reference models that can be used for comparison with feyn models.

class ConstantModel

def __init__(
    output_name,
    const
) -> ConstantModel

method ConstantModel.absolute_error

def absolute_error(
    self,
    data: Iterable
)

Compute the model's absolute error on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.absolute_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method ConstantModel.accuracy_score

def accuracy_score(
    self,
    data: Iterable
)

Compute the model's accuracy score on a data set.

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Formally it is defned as

(number of correct predictions) / (total number of preditions)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    accuracy score for the predictions

method ConstantModel.accuracy_threshold

def accuracy_threshold(
    self,
    data: Iterable
)

Compute the accuracy score of predictions with optimal threshold

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Accuracy is normally calculated under the assumption that the threshold that separates true from false is 0.5. Hovever, this is not the case when a model was trained with another population composition than on the one which is used.

This function first computes the threshold limining true from false classes that optimises the accuracy. It then returns this threshold along with the accuracy that is obtained using it.

Arguments:
    true -- Expected values
    pred -- Predicted values

Returns a tuple with:
    threshold that maximizes accuracy
    accuracy score obtained with this threshold

method ConstantModel.binary_cross_entropy

def binary_cross_entropy(
    self,
    data: Iterable
)

Compute the model's binary cross entropy on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.binary_cross_entropy(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method ConstantModel.mae

def mae(
    self,
    data
)

Compute the model's mean absolute error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MAE for the predictions

method ConstantModel.mse

def mse(
    self,
    data
)

Compute the model's mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MSE for the predictions

method ConstantModel.plot_confusion_matrix

def plot_confusion_matrix(
    self,
    data: Iterable,
    threshold: float = 0.5,
    labels: Iterable = None,
    title: str = 'Confusion matrix',
    color_map='feyn-primary',
    ax=None
) -> None

Compute and plot a Confusion Matrix.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Boundary of True and False predictions, default 0.5
    labels -- List of labels to index the matrix
    title -- Title of the plot.
    color_map -- Color map from matplotlib to use for the matrix
    ax -- matplotlib axes object to draw to, default None

method ConstantModel.plot_partial

def plot_partial(
    self,
    data: Iterable,
    by: str,
    fixed: Union[dict, NoneType] = None
) -> None

Plot a partial dependence plot.
This plot is useful to interpret the effect of a specific feature on the model output.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots._partial_dependence.plot_partial(best, data, by="age")

You can use any column in the dataset as the `by` parameter.
If you use a numerical column, the feature will vary from min to max of that varialbe in the training set.
If you use a categorical column, the feature will display all categories, sorted by the average prediction of that category.

Arguments:
    model -- The model to plot.
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to interpret by.
    fixed -- A dictionary of features and associated values to hold fixed

method ConstantModel.plot_probability_scores

def plot_probability_scores(
    self,
    data: Iterable,
    title='',
    nbins=10,
    h_args=None,
    ax=None
)

Plots the histogram of probability scores in binary
classification problems, highlighting the negative and
positive classes. Order of truth and prediction matters.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
Keyword Arguments:
    title {str} -- plot title (default: {''})
    nbins {int} -- number of bins (default: {10})
    h_args {dict} -- histogram kwargs (default: {None})
    ax {matplotlib.axes._subplots.AxesSubplot} -- axes object (default: {None})

method ConstantModel.plot_regression

def plot_regression(
    self,
    data: Iterable,
    title: str = 'Actuals vs Prediction',
    ax=None
)

This plots the true values on the x-axis and the predicted values on the y-axis.
On top of the plot is the line of equality y=x.
The closer the scattered values are to the line the better the predictions.
The line of best fit between y_true and y_pred is also calculated and plotted. This line should be close to the line y=x

Arguments:
    data {typing.Iterable} -- The dataset to determine regression quality. It contains input names and output name of the model as columns

Keyword Arguments:
    title {str} -- (default: {"Actuals vs Predictions"})
    ax {AxesSubplot} -- (default: {None})

method ConstantModel.plot_residuals

def plot_residuals(
    self,
    data: Iterable,
    title: str = 'Residuals plot',
    ax=None
)

This plots the predicted values against the residuals (y_true - y_pred).

Arguments:
    y_true {typing.Iterable} -- True values
    y_pred {typing.Iterable} -- Predicted values

Keyword Arguments:
    title {str} -- (default: {"Residual plot"})
    ax {[type]} -- (default: {None})

method ConstantModel.plot_roc_curve

def plot_roc_curve(
    self,
    data: Iterable,
    threshold: float = None,
    title: str = 'ROC curve',
    ax=None,
    **kwargs
) -> None

Plot the model's ROC curve.

This is a shorthand for calling feyn.plots.plot_roc_curve.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Plots a point on the ROC curve of the true positive rate and false positive rate at the given threshold. Default is None
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to, default None
    **kwargs -- additional options to pass on to matplotlib

method ConstantModel.plot_segmented_loss

def plot_segmented_loss(
    self,
    data: Iterable,
    by: Union[str, NoneType] = None,
    loss_function='squared_error',
    title='Segmented Loss',
    ax=None
) -> None

Plot the loss by segment of a dataset.

This plot is useful to evaluate how a model performs on different subsets of the data.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots.plot_segmented_loss(best, data, by="smoker")

This will plot a histogram of the model loss for smokers and non-smokers separately, which can help evaluate wheter the model has better performance for euther of the smoker sub-populations.

You can use any column in the dataset as the `by` parameter. If you use a numerical column, the data will be binned automatically.

Arguments:
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to segment by.
    loss_function -- The loss function to compute for each segmnent,
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to

method ConstantModel.predict

def predict(
    self,
    data: Iterable
)

method ConstantModel.r2_score

def r2_score(
    self,
    data: Iterable
)

Compute the model's r2 score on a data set

The r2 score for a regression model is defined as
1 - rss/tss

Where rss is the residual sum of squares for the predictions, and tss is the total sum of squares.
Intutively, the tss is the resuduals of a so-called "worst" model that always predicts the mean. Therefore, the r2 score expresses how much better the predictions are than such a model.

A result of 0 means that the model is no better than a model that always predicts the mean value
A result of 1 means that the model perfectly predicts the true value

It is possible to get r2 scores below 0 if the predictions are even worse than the mean model.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    r2 score for the predictions

method ConstantModel.rmse

def rmse(
    self,
    data
)

Compute the model's root mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    RMSE for the predictions

method ConstantModel.roc_auc_score

def roc_auc_score(
    self,
    data: Iterable
)

Calculate the Area Under Curve (AUC) of the ROC curve.

A ROC curve depicts the ability of a binary classifier with varying threshold.

The area under the curve (AUC) is the probability that said classifier will
attach a higher score to a random positive instance in comparison to a random
negative instance.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    AUC score for the predictions

method ConstantModel.squared_error

def squared_error(
    self,
    data: Iterable
)

Compute the model's squared error loss on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.squared_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

class GradientBoostingClassifier

def __init__(
    data,
    output_name,
    **kwargs
) -> GradientBoostingClassifier

method GradientBoostingClassifier.absolute_error

def absolute_error(
    self,
    data: Iterable
)

Compute the model's absolute error on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.absolute_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method GradientBoostingClassifier.accuracy_score

def accuracy_score(
    self,
    data: Iterable
)

Compute the model's accuracy score on a data set.

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Formally it is defned as

(number of correct predictions) / (total number of preditions)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    accuracy score for the predictions

method GradientBoostingClassifier.accuracy_threshold

def accuracy_threshold(
    self,
    data: Iterable
)

Compute the accuracy score of predictions with optimal threshold

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Accuracy is normally calculated under the assumption that the threshold that separates true from false is 0.5. Hovever, this is not the case when a model was trained with another population composition than on the one which is used.

This function first computes the threshold limining true from false classes that optimises the accuracy. It then returns this threshold along with the accuracy that is obtained using it.

Arguments:
    true -- Expected values
    pred -- Predicted values

Returns a tuple with:
    threshold that maximizes accuracy
    accuracy score obtained with this threshold

method GradientBoostingClassifier.binary_cross_entropy

def binary_cross_entropy(
    self,
    data: Iterable
)

Compute the model's binary cross entropy on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.binary_cross_entropy(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method GradientBoostingClassifier.mae

def mae(
    self,
    data
)

Compute the model's mean absolute error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MAE for the predictions

method GradientBoostingClassifier.mse

def mse(
    self,
    data
)

Compute the model's mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MSE for the predictions

method GradientBoostingClassifier.plot_confusion_matrix

def plot_confusion_matrix(
    self,
    data: Iterable,
    threshold: float = 0.5,
    labels: Iterable = None,
    title: str = 'Confusion matrix',
    color_map='feyn-primary',
    ax=None
) -> None

Compute and plot a Confusion Matrix.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Boundary of True and False predictions, default 0.5
    labels -- List of labels to index the matrix
    title -- Title of the plot.
    color_map -- Color map from matplotlib to use for the matrix
    ax -- matplotlib axes object to draw to, default None

method GradientBoostingClassifier.plot_partial

def plot_partial(
    self,
    data: Iterable,
    by: str,
    fixed: Union[dict, NoneType] = None
) -> None

Plot a partial dependence plot.
This plot is useful to interpret the effect of a specific feature on the model output.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots._partial_dependence.plot_partial(best, data, by="age")

You can use any column in the dataset as the `by` parameter.
If you use a numerical column, the feature will vary from min to max of that varialbe in the training set.
If you use a categorical column, the feature will display all categories, sorted by the average prediction of that category.

Arguments:
    model -- The model to plot.
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to interpret by.
    fixed -- A dictionary of features and associated values to hold fixed

method GradientBoostingClassifier.plot_probability_scores

def plot_probability_scores(
    self,
    data: Iterable,
    title='',
    nbins=10,
    h_args=None,
    ax=None
)

Plots the histogram of probability scores in binary
classification problems, highlighting the negative and
positive classes. Order of truth and prediction matters.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
Keyword Arguments:
    title {str} -- plot title (default: {''})
    nbins {int} -- number of bins (default: {10})
    h_args {dict} -- histogram kwargs (default: {None})
    ax {matplotlib.axes._subplots.AxesSubplot} -- axes object (default: {None})

method GradientBoostingClassifier.plot_regression

def plot_regression(
    self,
    data: Iterable,
    title: str = 'Actuals vs Prediction',
    ax=None
)

This plots the true values on the x-axis and the predicted values on the y-axis.
On top of the plot is the line of equality y=x.
The closer the scattered values are to the line the better the predictions.
The line of best fit between y_true and y_pred is also calculated and plotted. This line should be close to the line y=x

Arguments:
    data {typing.Iterable} -- The dataset to determine regression quality. It contains input names and output name of the model as columns

Keyword Arguments:
    title {str} -- (default: {"Actuals vs Predictions"})
    ax {AxesSubplot} -- (default: {None})

method GradientBoostingClassifier.plot_residuals

def plot_residuals(
    self,
    data: Iterable,
    title: str = 'Residuals plot',
    ax=None
)

This plots the predicted values against the residuals (y_true - y_pred).

Arguments:
    y_true {typing.Iterable} -- True values
    y_pred {typing.Iterable} -- Predicted values

Keyword Arguments:
    title {str} -- (default: {"Residual plot"})
    ax {[type]} -- (default: {None})

method GradientBoostingClassifier.plot_roc_curve

def plot_roc_curve(
    self,
    data: Iterable,
    threshold: float = None,
    title: str = 'ROC curve',
    ax=None,
    **kwargs
) -> None

Plot the model's ROC curve.

This is a shorthand for calling feyn.plots.plot_roc_curve.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Plots a point on the ROC curve of the true positive rate and false positive rate at the given threshold. Default is None
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to, default None
    **kwargs -- additional options to pass on to matplotlib

method GradientBoostingClassifier.plot_segmented_loss

def plot_segmented_loss(
    self,
    data: Iterable,
    by: Union[str, NoneType] = None,
    loss_function='squared_error',
    title='Segmented Loss',
    ax=None
) -> None

Plot the loss by segment of a dataset.

This plot is useful to evaluate how a model performs on different subsets of the data.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots.plot_segmented_loss(best, data, by="smoker")

This will plot a histogram of the model loss for smokers and non-smokers separately, which can help evaluate wheter the model has better performance for euther of the smoker sub-populations.

You can use any column in the dataset as the `by` parameter. If you use a numerical column, the data will be binned automatically.

Arguments:
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to segment by.
    loss_function -- The loss function to compute for each segmnent,
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to

method GradientBoostingClassifier.predict

def predict(
    self,
    X: Iterable
)

method GradientBoostingClassifier.r2_score

def r2_score(
    self,
    data: Iterable
)

Compute the model's r2 score on a data set

The r2 score for a regression model is defined as
1 - rss/tss

Where rss is the residual sum of squares for the predictions, and tss is the total sum of squares.
Intutively, the tss is the resuduals of a so-called "worst" model that always predicts the mean. Therefore, the r2 score expresses how much better the predictions are than such a model.

A result of 0 means that the model is no better than a model that always predicts the mean value
A result of 1 means that the model perfectly predicts the true value

It is possible to get r2 scores below 0 if the predictions are even worse than the mean model.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    r2 score for the predictions

method GradientBoostingClassifier.rmse

def rmse(
    self,
    data
)

Compute the model's root mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    RMSE for the predictions

method GradientBoostingClassifier.roc_auc_score

def roc_auc_score(
    self,
    data: Iterable
)

Calculate the Area Under Curve (AUC) of the ROC curve.

A ROC curve depicts the ability of a binary classifier with varying threshold.

The area under the curve (AUC) is the probability that said classifier will
attach a higher score to a random positive instance in comparison to a random
negative instance.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    AUC score for the predictions

method GradientBoostingClassifier.squared_error

def squared_error(
    self,
    data: Iterable
)

Compute the model's squared error loss on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.squared_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

class LinearRegression

def __init__(
    data,
    output_name,
    **kwargs
) -> LinearRegression

method LinearRegression.absolute_error

def absolute_error(
    self,
    data: Iterable
)

Compute the model's absolute error on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.absolute_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method LinearRegression.accuracy_score

def accuracy_score(
    self,
    data: Iterable
)

Compute the model's accuracy score on a data set.

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Formally it is defned as

(number of correct predictions) / (total number of preditions)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    accuracy score for the predictions

method LinearRegression.accuracy_threshold

def accuracy_threshold(
    self,
    data: Iterable
)

Compute the accuracy score of predictions with optimal threshold

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Accuracy is normally calculated under the assumption that the threshold that separates true from false is 0.5. Hovever, this is not the case when a model was trained with another population composition than on the one which is used.

This function first computes the threshold limining true from false classes that optimises the accuracy. It then returns this threshold along with the accuracy that is obtained using it.

Arguments:
    true -- Expected values
    pred -- Predicted values

Returns a tuple with:
    threshold that maximizes accuracy
    accuracy score obtained with this threshold

method LinearRegression.binary_cross_entropy

def binary_cross_entropy(
    self,
    data: Iterable
)

Compute the model's binary cross entropy on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.binary_cross_entropy(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method LinearRegression.mae

def mae(
    self,
    data
)

Compute the model's mean absolute error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MAE for the predictions

method LinearRegression.mse

def mse(
    self,
    data
)

Compute the model's mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MSE for the predictions

method LinearRegression.plot_confusion_matrix

def plot_confusion_matrix(
    self,
    data: Iterable,
    threshold: float = 0.5,
    labels: Iterable = None,
    title: str = 'Confusion matrix',
    color_map='feyn-primary',
    ax=None
) -> None

Compute and plot a Confusion Matrix.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Boundary of True and False predictions, default 0.5
    labels -- List of labels to index the matrix
    title -- Title of the plot.
    color_map -- Color map from matplotlib to use for the matrix
    ax -- matplotlib axes object to draw to, default None

method LinearRegression.plot_partial

def plot_partial(
    self,
    data: Iterable,
    by: str,
    fixed: Union[dict, NoneType] = None
) -> None

Plot a partial dependence plot.
This plot is useful to interpret the effect of a specific feature on the model output.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots._partial_dependence.plot_partial(best, data, by="age")

You can use any column in the dataset as the `by` parameter.
If you use a numerical column, the feature will vary from min to max of that varialbe in the training set.
If you use a categorical column, the feature will display all categories, sorted by the average prediction of that category.

Arguments:
    model -- The model to plot.
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to interpret by.
    fixed -- A dictionary of features and associated values to hold fixed

method LinearRegression.plot_probability_scores

def plot_probability_scores(
    self,
    data: Iterable,
    title='',
    nbins=10,
    h_args=None,
    ax=None
)

Plots the histogram of probability scores in binary
classification problems, highlighting the negative and
positive classes. Order of truth and prediction matters.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
Keyword Arguments:
    title {str} -- plot title (default: {''})
    nbins {int} -- number of bins (default: {10})
    h_args {dict} -- histogram kwargs (default: {None})
    ax {matplotlib.axes._subplots.AxesSubplot} -- axes object (default: {None})

method LinearRegression.plot_regression

def plot_regression(
    self,
    data: Iterable,
    title: str = 'Actuals vs Prediction',
    ax=None
)

This plots the true values on the x-axis and the predicted values on the y-axis.
On top of the plot is the line of equality y=x.
The closer the scattered values are to the line the better the predictions.
The line of best fit between y_true and y_pred is also calculated and plotted. This line should be close to the line y=x

Arguments:
    data {typing.Iterable} -- The dataset to determine regression quality. It contains input names and output name of the model as columns

Keyword Arguments:
    title {str} -- (default: {"Actuals vs Predictions"})
    ax {AxesSubplot} -- (default: {None})

method LinearRegression.plot_residuals

def plot_residuals(
    self,
    data: Iterable,
    title: str = 'Residuals plot',
    ax=None
)

This plots the predicted values against the residuals (y_true - y_pred).

Arguments:
    y_true {typing.Iterable} -- True values
    y_pred {typing.Iterable} -- Predicted values

Keyword Arguments:
    title {str} -- (default: {"Residual plot"})
    ax {[type]} -- (default: {None})

method LinearRegression.plot_roc_curve

def plot_roc_curve(
    self,
    data: Iterable,
    threshold: float = None,
    title: str = 'ROC curve',
    ax=None,
    **kwargs
) -> None

Plot the model's ROC curve.

This is a shorthand for calling feyn.plots.plot_roc_curve.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Plots a point on the ROC curve of the true positive rate and false positive rate at the given threshold. Default is None
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to, default None
    **kwargs -- additional options to pass on to matplotlib

method LinearRegression.plot_segmented_loss

def plot_segmented_loss(
    self,
    data: Iterable,
    by: Union[str, NoneType] = None,
    loss_function='squared_error',
    title='Segmented Loss',
    ax=None
) -> None

Plot the loss by segment of a dataset.

This plot is useful to evaluate how a model performs on different subsets of the data.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots.plot_segmented_loss(best, data, by="smoker")

This will plot a histogram of the model loss for smokers and non-smokers separately, which can help evaluate wheter the model has better performance for euther of the smoker sub-populations.

You can use any column in the dataset as the `by` parameter. If you use a numerical column, the data will be binned automatically.

Arguments:
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to segment by.
    loss_function -- The loss function to compute for each segmnent,
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to

method LinearRegression.predict

def predict(
    self,
    X: Iterable
)

method LinearRegression.r2_score

def r2_score(
    self,
    data: Iterable
)

Compute the model's r2 score on a data set

The r2 score for a regression model is defined as
1 - rss/tss

Where rss is the residual sum of squares for the predictions, and tss is the total sum of squares.
Intutively, the tss is the resuduals of a so-called "worst" model that always predicts the mean. Therefore, the r2 score expresses how much better the predictions are than such a model.

A result of 0 means that the model is no better than a model that always predicts the mean value
A result of 1 means that the model perfectly predicts the true value

It is possible to get r2 scores below 0 if the predictions are even worse than the mean model.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    r2 score for the predictions

method LinearRegression.rmse

def rmse(
    self,
    data
)

Compute the model's root mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    RMSE for the predictions

method LinearRegression.roc_auc_score

def roc_auc_score(
    self,
    data: Iterable
)

Calculate the Area Under Curve (AUC) of the ROC curve.

A ROC curve depicts the ability of a binary classifier with varying threshold.

The area under the curve (AUC) is the probability that said classifier will
attach a higher score to a random positive instance in comparison to a random
negative instance.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    AUC score for the predictions

method LinearRegression.squared_error

def squared_error(
    self,
    data: Iterable
)

Compute the model's squared error loss on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.squared_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

class LogisticRegressionClassifier

def __init__(
    data,
    output_name,
    **kwargs
) -> LogisticRegressionClassifier

method LogisticRegressionClassifier.absolute_error

def absolute_error(
    self,
    data: Iterable
)

Compute the model's absolute error on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.absolute_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method LogisticRegressionClassifier.accuracy_score

def accuracy_score(
    self,
    data: Iterable
)

Compute the model's accuracy score on a data set.

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Formally it is defned as

(number of correct predictions) / (total number of preditions)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    accuracy score for the predictions

method LogisticRegressionClassifier.accuracy_threshold

def accuracy_threshold(
    self,
    data: Iterable
)

Compute the accuracy score of predictions with optimal threshold

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Accuracy is normally calculated under the assumption that the threshold that separates true from false is 0.5. Hovever, this is not the case when a model was trained with another population composition than on the one which is used.

This function first computes the threshold limining true from false classes that optimises the accuracy. It then returns this threshold along with the accuracy that is obtained using it.

Arguments:
    true -- Expected values
    pred -- Predicted values

Returns a tuple with:
    threshold that maximizes accuracy
    accuracy score obtained with this threshold

method LogisticRegressionClassifier.binary_cross_entropy

def binary_cross_entropy(
    self,
    data: Iterable
)

Compute the model's binary cross entropy on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.binary_cross_entropy(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method LogisticRegressionClassifier.mae

def mae(
    self,
    data
)

Compute the model's mean absolute error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MAE for the predictions

method LogisticRegressionClassifier.mse

def mse(
    self,
    data
)

Compute the model's mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MSE for the predictions

method LogisticRegressionClassifier.plot_confusion_matrix

def plot_confusion_matrix(
    self,
    data: Iterable,
    threshold: float = 0.5,
    labels: Iterable = None,
    title: str = 'Confusion matrix',
    color_map='feyn-primary',
    ax=None
) -> None

Compute and plot a Confusion Matrix.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Boundary of True and False predictions, default 0.5
    labels -- List of labels to index the matrix
    title -- Title of the plot.
    color_map -- Color map from matplotlib to use for the matrix
    ax -- matplotlib axes object to draw to, default None

method LogisticRegressionClassifier.plot_partial

def plot_partial(
    self,
    data: Iterable,
    by: str,
    fixed: Union[dict, NoneType] = None
) -> None

Plot a partial dependence plot.
This plot is useful to interpret the effect of a specific feature on the model output.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots._partial_dependence.plot_partial(best, data, by="age")

You can use any column in the dataset as the `by` parameter.
If you use a numerical column, the feature will vary from min to max of that varialbe in the training set.
If you use a categorical column, the feature will display all categories, sorted by the average prediction of that category.

Arguments:
    model -- The model to plot.
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to interpret by.
    fixed -- A dictionary of features and associated values to hold fixed

method LogisticRegressionClassifier.plot_probability_scores

def plot_probability_scores(
    self,
    data: Iterable,
    title='',
    nbins=10,
    h_args=None,
    ax=None
)

Plots the histogram of probability scores in binary
classification problems, highlighting the negative and
positive classes. Order of truth and prediction matters.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
Keyword Arguments:
    title {str} -- plot title (default: {''})
    nbins {int} -- number of bins (default: {10})
    h_args {dict} -- histogram kwargs (default: {None})
    ax {matplotlib.axes._subplots.AxesSubplot} -- axes object (default: {None})

method LogisticRegressionClassifier.plot_regression

def plot_regression(
    self,
    data: Iterable,
    title: str = 'Actuals vs Prediction',
    ax=None
)

This plots the true values on the x-axis and the predicted values on the y-axis.
On top of the plot is the line of equality y=x.
The closer the scattered values are to the line the better the predictions.
The line of best fit between y_true and y_pred is also calculated and plotted. This line should be close to the line y=x

Arguments:
    data {typing.Iterable} -- The dataset to determine regression quality. It contains input names and output name of the model as columns

Keyword Arguments:
    title {str} -- (default: {"Actuals vs Predictions"})
    ax {AxesSubplot} -- (default: {None})

method LogisticRegressionClassifier.plot_residuals

def plot_residuals(
    self,
    data: Iterable,
    title: str = 'Residuals plot',
    ax=None
)

This plots the predicted values against the residuals (y_true - y_pred).

Arguments:
    y_true {typing.Iterable} -- True values
    y_pred {typing.Iterable} -- Predicted values

Keyword Arguments:
    title {str} -- (default: {"Residual plot"})
    ax {[type]} -- (default: {None})

method LogisticRegressionClassifier.plot_roc_curve

def plot_roc_curve(
    self,
    data: Iterable,
    threshold: float = None,
    title: str = 'ROC curve',
    ax=None,
    **kwargs
) -> None

Plot the model's ROC curve.

This is a shorthand for calling feyn.plots.plot_roc_curve.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Plots a point on the ROC curve of the true positive rate and false positive rate at the given threshold. Default is None
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to, default None
    **kwargs -- additional options to pass on to matplotlib

method LogisticRegressionClassifier.plot_segmented_loss

def plot_segmented_loss(
    self,
    data: Iterable,
    by: Union[str, NoneType] = None,
    loss_function='squared_error',
    title='Segmented Loss',
    ax=None
) -> None

Plot the loss by segment of a dataset.

This plot is useful to evaluate how a model performs on different subsets of the data.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots.plot_segmented_loss(best, data, by="smoker")

This will plot a histogram of the model loss for smokers and non-smokers separately, which can help evaluate wheter the model has better performance for euther of the smoker sub-populations.

You can use any column in the dataset as the `by` parameter. If you use a numerical column, the data will be binned automatically.

Arguments:
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to segment by.
    loss_function -- The loss function to compute for each segmnent,
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to

method LogisticRegressionClassifier.predict

def predict(
    self,
    X: Iterable
)

method LogisticRegressionClassifier.r2_score

def r2_score(
    self,
    data: Iterable
)

Compute the model's r2 score on a data set

The r2 score for a regression model is defined as
1 - rss/tss

Where rss is the residual sum of squares for the predictions, and tss is the total sum of squares.
Intutively, the tss is the resuduals of a so-called "worst" model that always predicts the mean. Therefore, the r2 score expresses how much better the predictions are than such a model.

A result of 0 means that the model is no better than a model that always predicts the mean value
A result of 1 means that the model perfectly predicts the true value

It is possible to get r2 scores below 0 if the predictions are even worse than the mean model.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    r2 score for the predictions

method LogisticRegressionClassifier.rmse

def rmse(
    self,
    data
)

Compute the model's root mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    RMSE for the predictions

method LogisticRegressionClassifier.roc_auc_score

def roc_auc_score(
    self,
    data: Iterable
)

Calculate the Area Under Curve (AUC) of the ROC curve.

A ROC curve depicts the ability of a binary classifier with varying threshold.

The area under the curve (AUC) is the probability that said classifier will
attach a higher score to a random positive instance in comparison to a random
negative instance.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    AUC score for the predictions

method LogisticRegressionClassifier.squared_error

def squared_error(
    self,
    data: Iterable
)

Compute the model's squared error loss on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.squared_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method LogisticRegressionClassifier.summary

def summary(
    self,
    ax=None
)

class RandomForestClassifier

def __init__(
    data,
    output_name,
    **kwargs
) -> RandomForestClassifier

method RandomForestClassifier.absolute_error

def absolute_error(
    self,
    data: Iterable
)

Compute the model's absolute error on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.absolute_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method RandomForestClassifier.accuracy_score

def accuracy_score(
    self,
    data: Iterable
)

Compute the model's accuracy score on a data set.

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Formally it is defned as

(number of correct predictions) / (total number of preditions)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    accuracy score for the predictions

method RandomForestClassifier.accuracy_threshold

def accuracy_threshold(
    self,
    data: Iterable
)

Compute the accuracy score of predictions with optimal threshold

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Accuracy is normally calculated under the assumption that the threshold that separates true from false is 0.5. Hovever, this is not the case when a model was trained with another population composition than on the one which is used.

This function first computes the threshold limining true from false classes that optimises the accuracy. It then returns this threshold along with the accuracy that is obtained using it.

Arguments:
    true -- Expected values
    pred -- Predicted values

Returns a tuple with:
    threshold that maximizes accuracy
    accuracy score obtained with this threshold

method RandomForestClassifier.binary_cross_entropy

def binary_cross_entropy(
    self,
    data: Iterable
)

Compute the model's binary cross entropy on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.binary_cross_entropy(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method RandomForestClassifier.mae

def mae(
    self,
    data
)

Compute the model's mean absolute error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MAE for the predictions

method RandomForestClassifier.mse

def mse(
    self,
    data
)

Compute the model's mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MSE for the predictions

method RandomForestClassifier.plot_confusion_matrix

def plot_confusion_matrix(
    self,
    data: Iterable,
    threshold: float = 0.5,
    labels: Iterable = None,
    title: str = 'Confusion matrix',
    color_map='feyn-primary',
    ax=None
) -> None

Compute and plot a Confusion Matrix.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Boundary of True and False predictions, default 0.5
    labels -- List of labels to index the matrix
    title -- Title of the plot.
    color_map -- Color map from matplotlib to use for the matrix
    ax -- matplotlib axes object to draw to, default None

method RandomForestClassifier.plot_partial

def plot_partial(
    self,
    data: Iterable,
    by: str,
    fixed: Union[dict, NoneType] = None
) -> None

Plot a partial dependence plot.
This plot is useful to interpret the effect of a specific feature on the model output.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots._partial_dependence.plot_partial(best, data, by="age")

You can use any column in the dataset as the `by` parameter.
If you use a numerical column, the feature will vary from min to max of that varialbe in the training set.
If you use a categorical column, the feature will display all categories, sorted by the average prediction of that category.

Arguments:
    model -- The model to plot.
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to interpret by.
    fixed -- A dictionary of features and associated values to hold fixed

method RandomForestClassifier.plot_probability_scores

def plot_probability_scores(
    self,
    data: Iterable,
    title='',
    nbins=10,
    h_args=None,
    ax=None
)

Plots the histogram of probability scores in binary
classification problems, highlighting the negative and
positive classes. Order of truth and prediction matters.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
Keyword Arguments:
    title {str} -- plot title (default: {''})
    nbins {int} -- number of bins (default: {10})
    h_args {dict} -- histogram kwargs (default: {None})
    ax {matplotlib.axes._subplots.AxesSubplot} -- axes object (default: {None})

method RandomForestClassifier.plot_regression

def plot_regression(
    self,
    data: Iterable,
    title: str = 'Actuals vs Prediction',
    ax=None
)

This plots the true values on the x-axis and the predicted values on the y-axis.
On top of the plot is the line of equality y=x.
The closer the scattered values are to the line the better the predictions.
The line of best fit between y_true and y_pred is also calculated and plotted. This line should be close to the line y=x

Arguments:
    data {typing.Iterable} -- The dataset to determine regression quality. It contains input names and output name of the model as columns

Keyword Arguments:
    title {str} -- (default: {"Actuals vs Predictions"})
    ax {AxesSubplot} -- (default: {None})

method RandomForestClassifier.plot_residuals

def plot_residuals(
    self,
    data: Iterable,
    title: str = 'Residuals plot',
    ax=None
)

This plots the predicted values against the residuals (y_true - y_pred).

Arguments:
    y_true {typing.Iterable} -- True values
    y_pred {typing.Iterable} -- Predicted values

Keyword Arguments:
    title {str} -- (default: {"Residual plot"})
    ax {[type]} -- (default: {None})

method RandomForestClassifier.plot_roc_curve

def plot_roc_curve(
    self,
    data: Iterable,
    threshold: float = None,
    title: str = 'ROC curve',
    ax=None,
    **kwargs
) -> None

Plot the model's ROC curve.

This is a shorthand for calling feyn.plots.plot_roc_curve.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Plots a point on the ROC curve of the true positive rate and false positive rate at the given threshold. Default is None
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to, default None
    **kwargs -- additional options to pass on to matplotlib

method RandomForestClassifier.plot_segmented_loss

def plot_segmented_loss(
    self,
    data: Iterable,
    by: Union[str, NoneType] = None,
    loss_function='squared_error',
    title='Segmented Loss',
    ax=None
) -> None

Plot the loss by segment of a dataset.

This plot is useful to evaluate how a model performs on different subsets of the data.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots.plot_segmented_loss(best, data, by="smoker")

This will plot a histogram of the model loss for smokers and non-smokers separately, which can help evaluate wheter the model has better performance for euther of the smoker sub-populations.

You can use any column in the dataset as the `by` parameter. If you use a numerical column, the data will be binned automatically.

Arguments:
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to segment by.
    loss_function -- The loss function to compute for each segmnent,
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to

method RandomForestClassifier.predict

def predict(
    self,
    X: Iterable
)

method RandomForestClassifier.r2_score

def r2_score(
    self,
    data: Iterable
)

Compute the model's r2 score on a data set

The r2 score for a regression model is defined as
1 - rss/tss

Where rss is the residual sum of squares for the predictions, and tss is the total sum of squares.
Intutively, the tss is the resuduals of a so-called "worst" model that always predicts the mean. Therefore, the r2 score expresses how much better the predictions are than such a model.

A result of 0 means that the model is no better than a model that always predicts the mean value
A result of 1 means that the model perfectly predicts the true value

It is possible to get r2 scores below 0 if the predictions are even worse than the mean model.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    r2 score for the predictions

method RandomForestClassifier.rmse

def rmse(
    self,
    data
)

Compute the model's root mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    RMSE for the predictions

method RandomForestClassifier.roc_auc_score

def roc_auc_score(
    self,
    data: Iterable
)

Calculate the Area Under Curve (AUC) of the ROC curve.

A ROC curve depicts the ability of a binary classifier with varying threshold.

The area under the curve (AUC) is the probability that said classifier will
attach a higher score to a random positive instance in comparison to a random
negative instance.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    AUC score for the predictions

method RandomForestClassifier.squared_error

def squared_error(
    self,
    data: Iterable
)

Compute the model's squared error loss on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.squared_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

class SKLeanClassifier

def __init__(
    sklearn_classifier: type,
    data,
    output_name,
    **kwargs
) -> SKLeanClassifier

method SKLeanClassifier.absolute_error

def absolute_error(
    self,
    data: Iterable
)

Compute the model's absolute error on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.absolute_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method SKLeanClassifier.accuracy_score

def accuracy_score(
    self,
    data: Iterable
)

Compute the model's accuracy score on a data set.

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Formally it is defned as

(number of correct predictions) / (total number of preditions)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    accuracy score for the predictions

method SKLeanClassifier.accuracy_threshold

def accuracy_threshold(
    self,
    data: Iterable
)

Compute the accuracy score of predictions with optimal threshold

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Accuracy is normally calculated under the assumption that the threshold that separates true from false is 0.5. Hovever, this is not the case when a model was trained with another population composition than on the one which is used.

This function first computes the threshold limining true from false classes that optimises the accuracy. It then returns this threshold along with the accuracy that is obtained using it.

Arguments:
    true -- Expected values
    pred -- Predicted values

Returns a tuple with:
    threshold that maximizes accuracy
    accuracy score obtained with this threshold

method SKLeanClassifier.binary_cross_entropy

def binary_cross_entropy(
    self,
    data: Iterable
)

Compute the model's binary cross entropy on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.binary_cross_entropy(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method SKLeanClassifier.mae

def mae(
    self,
    data
)

Compute the model's mean absolute error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MAE for the predictions

method SKLeanClassifier.mse

def mse(
    self,
    data
)

Compute the model's mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MSE for the predictions

method SKLeanClassifier.plot_confusion_matrix

def plot_confusion_matrix(
    self,
    data: Iterable,
    threshold: float = 0.5,
    labels: Iterable = None,
    title: str = 'Confusion matrix',
    color_map='feyn-primary',
    ax=None
) -> None

Compute and plot a Confusion Matrix.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Boundary of True and False predictions, default 0.5
    labels -- List of labels to index the matrix
    title -- Title of the plot.
    color_map -- Color map from matplotlib to use for the matrix
    ax -- matplotlib axes object to draw to, default None

method SKLeanClassifier.plot_partial

def plot_partial(
    self,
    data: Iterable,
    by: str,
    fixed: Union[dict, NoneType] = None
) -> None

Plot a partial dependence plot.
This plot is useful to interpret the effect of a specific feature on the model output.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots._partial_dependence.plot_partial(best, data, by="age")

You can use any column in the dataset as the `by` parameter.
If you use a numerical column, the feature will vary from min to max of that varialbe in the training set.
If you use a categorical column, the feature will display all categories, sorted by the average prediction of that category.

Arguments:
    model -- The model to plot.
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to interpret by.
    fixed -- A dictionary of features and associated values to hold fixed

method SKLeanClassifier.plot_probability_scores

def plot_probability_scores(
    self,
    data: Iterable,
    title='',
    nbins=10,
    h_args=None,
    ax=None
)

Plots the histogram of probability scores in binary
classification problems, highlighting the negative and
positive classes. Order of truth and prediction matters.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
Keyword Arguments:
    title {str} -- plot title (default: {''})
    nbins {int} -- number of bins (default: {10})
    h_args {dict} -- histogram kwargs (default: {None})
    ax {matplotlib.axes._subplots.AxesSubplot} -- axes object (default: {None})

method SKLeanClassifier.plot_regression

def plot_regression(
    self,
    data: Iterable,
    title: str = 'Actuals vs Prediction',
    ax=None
)

This plots the true values on the x-axis and the predicted values on the y-axis.
On top of the plot is the line of equality y=x.
The closer the scattered values are to the line the better the predictions.
The line of best fit between y_true and y_pred is also calculated and plotted. This line should be close to the line y=x

Arguments:
    data {typing.Iterable} -- The dataset to determine regression quality. It contains input names and output name of the model as columns

Keyword Arguments:
    title {str} -- (default: {"Actuals vs Predictions"})
    ax {AxesSubplot} -- (default: {None})

method SKLeanClassifier.plot_residuals

def plot_residuals(
    self,
    data: Iterable,
    title: str = 'Residuals plot',
    ax=None
)

This plots the predicted values against the residuals (y_true - y_pred).

Arguments:
    y_true {typing.Iterable} -- True values
    y_pred {typing.Iterable} -- Predicted values

Keyword Arguments:
    title {str} -- (default: {"Residual plot"})
    ax {[type]} -- (default: {None})

method SKLeanClassifier.plot_roc_curve

def plot_roc_curve(
    self,
    data: Iterable,
    threshold: float = None,
    title: str = 'ROC curve',
    ax=None,
    **kwargs
) -> None

Plot the model's ROC curve.

This is a shorthand for calling feyn.plots.plot_roc_curve.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Plots a point on the ROC curve of the true positive rate and false positive rate at the given threshold. Default is None
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to, default None
    **kwargs -- additional options to pass on to matplotlib

method SKLeanClassifier.plot_segmented_loss

def plot_segmented_loss(
    self,
    data: Iterable,
    by: Union[str, NoneType] = None,
    loss_function='squared_error',
    title='Segmented Loss',
    ax=None
) -> None

Plot the loss by segment of a dataset.

This plot is useful to evaluate how a model performs on different subsets of the data.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots.plot_segmented_loss(best, data, by="smoker")

This will plot a histogram of the model loss for smokers and non-smokers separately, which can help evaluate wheter the model has better performance for euther of the smoker sub-populations.

You can use any column in the dataset as the `by` parameter. If you use a numerical column, the data will be binned automatically.

Arguments:
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to segment by.
    loss_function -- The loss function to compute for each segmnent,
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to

method SKLeanClassifier.predict

def predict(
    self,
    X: Iterable
)

method SKLeanClassifier.r2_score

def r2_score(
    self,
    data: Iterable
)

Compute the model's r2 score on a data set

The r2 score for a regression model is defined as
1 - rss/tss

Where rss is the residual sum of squares for the predictions, and tss is the total sum of squares.
Intutively, the tss is the resuduals of a so-called "worst" model that always predicts the mean. Therefore, the r2 score expresses how much better the predictions are than such a model.

A result of 0 means that the model is no better than a model that always predicts the mean value
A result of 1 means that the model perfectly predicts the true value

It is possible to get r2 scores below 0 if the predictions are even worse than the mean model.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    r2 score for the predictions

method SKLeanClassifier.rmse

def rmse(
    self,
    data
)

Compute the model's root mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    RMSE for the predictions

method SKLeanClassifier.roc_auc_score

def roc_auc_score(
    self,
    data: Iterable
)

Calculate the Area Under Curve (AUC) of the ROC curve.

A ROC curve depicts the ability of a binary classifier with varying threshold.

The area under the curve (AUC) is the probability that said classifier will
attach a higher score to a random positive instance in comparison to a random
negative instance.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    AUC score for the predictions

method SKLeanClassifier.squared_error

def squared_error(
    self,
    data: Iterable
)

Compute the model's squared error loss on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.squared_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

class SKLearnRegressor

def __init__(
    sklearn_regressor: type,
    data,
    output_name,
    **kwargs
) -> SKLearnRegressor

method SKLearnRegressor.absolute_error

def absolute_error(
    self,
    data: Iterable
)

Compute the model's absolute error on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.absolute_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method SKLearnRegressor.accuracy_score

def accuracy_score(
    self,
    data: Iterable
)

Compute the model's accuracy score on a data set.

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Formally it is defned as

(number of correct predictions) / (total number of preditions)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    accuracy score for the predictions

method SKLearnRegressor.accuracy_threshold

def accuracy_threshold(
    self,
    data: Iterable
)

Compute the accuracy score of predictions with optimal threshold

The accuracy score is useful to evaluate classification models. It is the fraction of the preditions that are correct. Accuracy is normally calculated under the assumption that the threshold that separates true from false is 0.5. Hovever, this is not the case when a model was trained with another population composition than on the one which is used.

This function first computes the threshold limining true from false classes that optimises the accuracy. It then returns this threshold along with the accuracy that is obtained using it.

Arguments:
    true -- Expected values
    pred -- Predicted values

Returns a tuple with:
    threshold that maximizes accuracy
    accuracy score obtained with this threshold

method SKLearnRegressor.binary_cross_entropy

def binary_cross_entropy(
    self,
    data: Iterable
)

Compute the model's binary cross entropy on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.binary_cross_entropy(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.

method SKLearnRegressor.mae

def mae(
    self,
    data
)

Compute the model's mean absolute error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MAE for the predictions

method SKLearnRegressor.mse

def mse(
    self,
    data
)

Compute the model's mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    MSE for the predictions

method SKLearnRegressor.plot_confusion_matrix

def plot_confusion_matrix(
    self,
    data: Iterable,
    threshold: float = 0.5,
    labels: Iterable = None,
    title: str = 'Confusion matrix',
    color_map='feyn-primary',
    ax=None
) -> None

Compute and plot a Confusion Matrix.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Boundary of True and False predictions, default 0.5
    labels -- List of labels to index the matrix
    title -- Title of the plot.
    color_map -- Color map from matplotlib to use for the matrix
    ax -- matplotlib axes object to draw to, default None

method SKLearnRegressor.plot_partial

def plot_partial(
    self,
    data: Iterable,
    by: str,
    fixed: Union[dict, NoneType] = None
) -> None

Plot a partial dependence plot.
This plot is useful to interpret the effect of a specific feature on the model output.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots._partial_dependence.plot_partial(best, data, by="age")

You can use any column in the dataset as the `by` parameter.
If you use a numerical column, the feature will vary from min to max of that varialbe in the training set.
If you use a categorical column, the feature will display all categories, sorted by the average prediction of that category.

Arguments:
    model -- The model to plot.
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to interpret by.
    fixed -- A dictionary of features and associated values to hold fixed

method SKLearnRegressor.plot_probability_scores

def plot_probability_scores(
    self,
    data: Iterable,
    title='',
    nbins=10,
    h_args=None,
    ax=None
)

Plots the histogram of probability scores in binary
classification problems, highlighting the negative and
positive classes. Order of truth and prediction matters.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
Keyword Arguments:
    title {str} -- plot title (default: {''})
    nbins {int} -- number of bins (default: {10})
    h_args {dict} -- histogram kwargs (default: {None})
    ax {matplotlib.axes._subplots.AxesSubplot} -- axes object (default: {None})

method SKLearnRegressor.plot_regression

def plot_regression(
    self,
    data: Iterable,
    title: str = 'Actuals vs Prediction',
    ax=None
)

This plots the true values on the x-axis and the predicted values on the y-axis.
On top of the plot is the line of equality y=x.
The closer the scattered values are to the line the better the predictions.
The line of best fit between y_true and y_pred is also calculated and plotted. This line should be close to the line y=x

Arguments:
    data {typing.Iterable} -- The dataset to determine regression quality. It contains input names and output name of the model as columns

Keyword Arguments:
    title {str} -- (default: {"Actuals vs Predictions"})
    ax {AxesSubplot} -- (default: {None})

method SKLearnRegressor.plot_residuals

def plot_residuals(
    self,
    data: Iterable,
    title: str = 'Residuals plot',
    ax=None
)

This plots the predicted values against the residuals (y_true - y_pred).

Arguments:
    y_true {typing.Iterable} -- True values
    y_pred {typing.Iterable} -- Predicted values

Keyword Arguments:
    title {str} -- (default: {"Residual plot"})
    ax {[type]} -- (default: {None})

method SKLearnRegressor.plot_roc_curve

def plot_roc_curve(
    self,
    data: Iterable,
    threshold: float = None,
    title: str = 'ROC curve',
    ax=None,
    **kwargs
) -> None

Plot the model's ROC curve.

This is a shorthand for calling feyn.plots.plot_roc_curve.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.
    threshold -- Plots a point on the ROC curve of the true positive rate and false positive rate at the given threshold. Default is None
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to, default None
    **kwargs -- additional options to pass on to matplotlib

method SKLearnRegressor.plot_segmented_loss

def plot_segmented_loss(
    self,
    data: Iterable,
    by: Union[str, NoneType] = None,
    loss_function='squared_error',
    title='Segmented Loss',
    ax=None
) -> None

Plot the loss by segment of a dataset.

This plot is useful to evaluate how a model performs on different subsets of the data.

Example:
> models = qlattice.sample_models(["age","smoker","heartrate"], output="heartrate")
> models = feyn.fit_models(models, data)
> best = models[0]
> feyn.plots.plot_segmented_loss(best, data, by="smoker")

This will plot a histogram of the model loss for smokers and non-smokers separately, which can help evaluate wheter the model has better performance for euther of the smoker sub-populations.

You can use any column in the dataset as the `by` parameter. If you use a numerical column, the data will be binned automatically.

Arguments:
    data -- The dataset to measure the loss on.
    by -- The column in the dataset to segment by.
    loss_function -- The loss function to compute for each segmnent,
    title -- Title of the plot.
    ax -- matplotlib axes object to draw to

method SKLearnRegressor.predict

def predict(
    self,
    X: Iterable
)

method SKLearnRegressor.r2_score

def r2_score(
    self,
    data: Iterable
)

Compute the model's r2 score on a data set

The r2 score for a regression model is defined as
1 - rss/tss

Where rss is the residual sum of squares for the predictions, and tss is the total sum of squares.
Intutively, the tss is the resuduals of a so-called "worst" model that always predicts the mean. Therefore, the r2 score expresses how much better the predictions are than such a model.

A result of 0 means that the model is no better than a model that always predicts the mean value
A result of 1 means that the model perfectly predicts the true value

It is possible to get r2 scores below 0 if the predictions are even worse than the mean model.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    r2 score for the predictions

method SKLearnRegressor.rmse

def rmse(
    self,
    data
)

Compute the model's root mean squared error on a data set.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    RMSE for the predictions

method SKLearnRegressor.roc_auc_score

def roc_auc_score(
    self,
    data: Iterable
)

Calculate the Area Under Curve (AUC) of the ROC curve.

A ROC curve depicts the ability of a binary classifier with varying threshold.

The area under the curve (AUC) is the probability that said classifier will
attach a higher score to a random positive instance in comparison to a random
negative instance.

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    AUC score for the predictions

method SKLearnRegressor.squared_error

def squared_error(
    self,
    data: Iterable
)

Compute the model's squared error loss on the provided data.

This function is a shorthand that is equivalent to the following code:
> y_true = data[
]
> y_pred = model.predict(data)
> se = feyn.losses.squared_error(y_true, y_pred)

Arguments:
    data -- Data set including both input and expected values. Can be either a dict mapping register names to value arrays, or a pandas.DataFrame.

Returns:
    nd.array -- The losses as an array of floats.