MLPClassifier
Multi-layer Perceptron classifier.
This model optimizes the log-loss function using LBFGS or stochastic gradient descent.
Python Reference (opens in a new tab)
Constructors
constructor()
Signature
new MLPClassifier(opts?: object): MLPClassifier;Parameters
| Name | Type | Description |
|---|---|---|
opts? | object | - |
opts.activation? | "identity" | "logistic" | "tanh" | "relu" | Activation function for the hidden layer. Default Value 'relu' |
opts.alpha? | number | Strength of the L2 regularization term. The L2 regularization term is divided by the sample size when added to the loss. Default Value 0.0001 |
opts.batch_size? | number | Size of minibatches for stochastic optimizers. If the solver is ‘lbfgs’, the classifier will not use minibatch. When set to “auto”, batch\_size=min(200, n\_samples). Default Value 'auto' |
opts.beta_1? | number | Exponential decay rate for estimates of first moment vector in adam, should be in \0, 1). Only used when solver=’adam’. Default Value 0.9 |
opts.beta_2? | number | Exponential decay rate for estimates of second moment vector in adam, should be in [0, 1). Only used when solver=’adam’. Default Value 0.999 |
opts.early_stopping? | boolean | Whether to use early stopping to terminate training when validation score is not improving. If set to true, it will automatically set aside 10% of training data as validation and terminate training when validation score is not improving by at least tol for n\_iter\_no\_change consecutive epochs. The split is stratified, except in a multilabel setting. If early stopping is false, then the training stops when the training loss does not improve by more than tol for n_iter_no_change consecutive passes over the training set. Only effective when solver=’sgd’ or ‘adam’. Default Value false |
opts.epsilon? | number | Value for numerical stability in adam. Only used when solver=’adam’. Default Value 1e-8 |
opts.hidden_layer_sizes? | any | The ith element represents the number of neurons in the ith hidden layer. |
opts.learning_rate? | "constant" | "invscaling" | "adaptive" | Learning rate schedule for weight updates. Default Value 'constant' |
opts.learning_rate_init? | number | The initial learning rate used. It controls the step-size in updating the weights. Only used when solver=’sgd’ or ‘adam’. Default Value 0.001 |
opts.max_fun? | number | Only used when solver=’lbfgs’. Maximum number of loss function calls. The solver iterates until convergence (determined by ‘tol’), number of iterations reaches max_iter, or this number of loss function calls. Note that number of loss function calls will be greater than or equal to the number of iterations for the MLPClassifier. Default Value 15000 |
opts.max_iter? | number | Maximum number of iterations. The solver iterates until convergence (determined by ‘tol’) or this number of iterations. For stochastic solvers (‘sgd’, ‘adam’), note that this determines the number of epochs (how many times each data point will be used), not the number of gradient steps. Default Value 200 |
opts.momentum? | number | Momentum for gradient descent update. Should be between 0 and 1. Only used when solver=’sgd’. Default Value 0.9 |
opts.n_iter_no_change? | number | Maximum number of epochs to not meet tol improvement. Only effective when solver=’sgd’ or ‘adam’. Default Value 10 |
opts.nesterovs_momentum? | boolean | Whether to use Nesterov’s momentum. Only used when solver=’sgd’ and momentum > 0. Default Value true |
opts.power_t? | number | The exponent for inverse scaling learning rate. It is used in updating effective learning rate when the learning_rate is set to ‘invscaling’. Only used when solver=’sgd’. Default Value 0.5 |
opts.random_state? | number | Determines random number generation for weights and bias initialization, train-test split if early stopping is used, and batch sampling when solver=’sgd’ or ‘adam’. Pass an int for reproducible results across multiple function calls. See [Glossary. |
opts.shuffle? | boolean | Whether to shuffle samples in each iteration. Only used when solver=’sgd’ or ‘adam’. Default Value true |
opts.solver? | "lbfgs" | "sgd" | "adam" | The solver for weight optimization. Default Value 'adam' |
opts.tol? | number | Tolerance for the optimization. When the loss or score is not improving by at least tol for n\_iter\_no\_change consecutive iterations, unless learning\_rate is set to ‘adaptive’, convergence is considered to be reached and training stops. Default Value 0.0001 |
opts.validation_fraction? | number | The proportion of training data to set aside as validation set for early stopping. Must be between 0 and 1. Only used if early_stopping is true. Default Value 0.1 |
opts.verbose? | boolean | Whether to print progress messages to stdout. Default Value false |
opts.warm_start? | boolean | When set to true, reuse the solution of the previous call to fit as initialization, otherwise, just erase the previous solution. See the Glossary. Default Value false |
Returns
Defined in: generated/neural_network/MLPClassifier.ts:23 (opens in a new tab)
Properties
_isDisposed
boolean=false
Defined in: generated/neural_network/MLPClassifier.ts:21 (opens in a new tab)
_isInitialized
boolean=false
Defined in: generated/neural_network/MLPClassifier.ts:20 (opens in a new tab)
_py
PythonBridge
Defined in: generated/neural_network/MLPClassifier.ts:19 (opens in a new tab)
id
string
Defined in: generated/neural_network/MLPClassifier.ts:16 (opens in a new tab)
opts
any
Defined in: generated/neural_network/MLPClassifier.ts:17 (opens in a new tab)
Accessors
best_loss_
The minimum loss reached by the solver throughout fitting. If early\_stopping=True, this attribute is set ot undefined. Refer to the best\_validation\_score\_ fitted attribute instead.
Signature
best_loss_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:570 (opens in a new tab)
best_validation_score_
The best validation score (i.e. accuracy score) that triggered the early stopping. Only available if early\_stopping=True, otherwise the attribute is set to undefined.
Signature
best_validation_score_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:645 (opens in a new tab)
classes_
Class labels for each output.
Signature
classes_(): Promise<ArrayLike>;Returns
Promise<ArrayLike>
Defined in: generated/neural_network/MLPClassifier.ts:522 (opens in a new tab)
coefs_
The ith element in the list represents the weight matrix corresponding to layer i.
Signature
coefs_(): Promise<any[]>;Returns
Promise<any[]>
Defined in: generated/neural_network/MLPClassifier.ts:693 (opens in a new tab)
feature_names_in_
Names of features seen during fit. Defined only when X has feature names that are all strings.
Signature
feature_names_in_(): Promise<ArrayLike>;Returns
Promise<ArrayLike>
Defined in: generated/neural_network/MLPClassifier.ts:766 (opens in a new tab)
intercepts_
The ith element in the list represents the bias vector corresponding to layer i + 1.
Signature
intercepts_(): Promise<any[]>;Returns
Promise<any[]>
Defined in: generated/neural_network/MLPClassifier.ts:716 (opens in a new tab)
loss_
The current loss computed with the loss function.
Signature
loss_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:547 (opens in a new tab)
loss_curve_
The ith element in the list represents the loss at the ith iteration.
Signature
loss_curve_(): Promise<any[]>;Returns
Promise<any[]>
Defined in: generated/neural_network/MLPClassifier.ts:595 (opens in a new tab)
n_features_in_
Number of features seen during fit.
Signature
n_features_in_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:741 (opens in a new tab)
n_iter_
The number of iterations the solver has run.
Signature
n_iter_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:791 (opens in a new tab)
n_layers_
Number of layers.
Signature
n_layers_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:814 (opens in a new tab)
n_outputs_
Number of outputs.
Signature
n_outputs_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:839 (opens in a new tab)
out_activation_
Name of the output activation function.
Signature
out_activation_(): Promise<string>;Returns
Promise<string>
Defined in: generated/neural_network/MLPClassifier.ts:864 (opens in a new tab)
py
Signature
py(): PythonBridge;Returns
PythonBridge
Defined in: generated/neural_network/MLPClassifier.ts:185 (opens in a new tab)
Signature
py(pythonBridge: PythonBridge): void;Parameters
| Name | Type |
|---|---|
pythonBridge | PythonBridge |
Returns
void
Defined in: generated/neural_network/MLPClassifier.ts:189 (opens in a new tab)
t_
The number of training samples seen by the solver during fitting.
Signature
t_(): Promise<number>;Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:670 (opens in a new tab)
validation_scores_
The score at each iteration on a held-out validation set. The score reported is the accuracy score. Only available if early\_stopping=True, otherwise the attribute is set to undefined.
Signature
validation_scores_(): Promise<any[]>;Returns
Promise<any[]>
Defined in: generated/neural_network/MLPClassifier.ts:620 (opens in a new tab)
Methods
dispose()
Disposes of the underlying Python resources.
Once dispose() is called, the instance is no longer usable.
Signature
dispose(): Promise<void>;Returns
Promise<void>
Defined in: generated/neural_network/MLPClassifier.ts:268 (opens in a new tab)
fit()
Fit the model to data matrix X and target(s) y.
Signature
fit(opts: object): Promise<any>;Parameters
| Name | Type | Description |
|---|---|---|
opts | object | - |
opts.X? | ArrayLike | The input data. |
opts.y? | ArrayLike | The target values (class labels in classification, real numbers in regression). |
Returns
Promise<any>
Defined in: generated/neural_network/MLPClassifier.ts:285 (opens in a new tab)
init()
Initializes the underlying Python resources.
This instance is not usable until the Promise returned by init() resolves.
Signature
init(py: PythonBridge): Promise<void>;Parameters
| Name | Type |
|---|---|
py | PythonBridge |
Returns
Promise<void>
Defined in: generated/neural_network/MLPClassifier.ts:198 (opens in a new tab)
partial_fit()
Update the model with a single iteration over the given data.
Signature
partial_fit(opts: object): Promise<any>;Parameters
| Name | Type | Description |
|---|---|---|
opts | object | - |
opts.X? | ArrayLike | The input data. |
opts.classes? | any[] | Classes across all calls to partial_fit. Can be obtained via np.unique(y\_all), where y_all is the target vector of the entire dataset. This argument is required for the first call to partial_fit and can be omitted in the subsequent calls. Note that y doesn’t need to contain all labels in classes. |
opts.y? | ArrayLike | The target values. |
Returns
Promise<any>
Defined in: generated/neural_network/MLPClassifier.ts:325 (opens in a new tab)
predict()
Predict using the multi-layer perceptron classifier.
Signature
predict(opts: object): Promise<ArrayLike>;Parameters
| Name | Type | Description |
|---|---|---|
opts | object | - |
opts.X? | ArrayLike | The input data. |
Returns
Promise<ArrayLike>
Defined in: generated/neural_network/MLPClassifier.ts:372 (opens in a new tab)
predict_log_proba()
Return the log of probability estimates.
Signature
predict_log_proba(opts: object): Promise<ArrayLike[]>;Parameters
| Name | Type | Description |
|---|---|---|
opts | object | - |
opts.X? | ArrayLike[] | The input data. |
Returns
Promise<ArrayLike[]>
Defined in: generated/neural_network/MLPClassifier.ts:405 (opens in a new tab)
predict_proba()
Probability estimates.
Signature
predict_proba(opts: object): Promise<ArrayLike[]>;Parameters
| Name | Type | Description |
|---|---|---|
opts | object | - |
opts.X? | ArrayLike | The input data. |
Returns
Promise<ArrayLike[]>
Defined in: generated/neural_network/MLPClassifier.ts:440 (opens in a new tab)
score()
Return the mean accuracy on the given test data and labels.
In multi-label classification, this is the subset accuracy which is a harsh metric since you require for each sample that each label set be correctly predicted.
Signature
score(opts: object): Promise<number>;Parameters
| Name | Type | Description |
|---|---|---|
opts | object | - |
opts.X? | ArrayLike[] | Test samples. |
opts.sample_weight? | ArrayLike | Sample weights. |
opts.y? | ArrayLike | True labels for X. |
Returns
Promise<number>
Defined in: generated/neural_network/MLPClassifier.ts:475 (opens in a new tab)