rocelib.models.keras_models package
Submodules
rocelib.models.keras_models.TrainableKerasModel module
- class rocelib.models.keras_models.TrainableKerasModel.TrainableKerasModel(input_dim, hidden_dim, output_dim)[source]
Bases:
TrainableModel
A simple feedforward neural network model using Keras for binary classification.
This model includes one hidden layer with ReLU activation and an output layer with a sigmoid activation function. It utilizes the Adam optimizer with a learning rate of 0.001 and the binary cross-entropy loss function.
- model
The Keras Sequential model instance containing the neural network architecture.
- Type:
keras.models.Sequential
- __init__(input_dim: int, hidden_dim: int, output_dim: int):
Initializes the neural network model with the specified dimensions.
- train(X: pd.DataFrame, y: pd.DataFrame, epochs: int = 100, batch_size: int = 32) None: [source]
Trains the model using the provided feature and target variables.
- predict_single(x: pd.DataFrame) int: [source]
Predicts the outcome for a single instance and returns the class label.
- evaluate(X: pd.DataFrame, y: pd.DataFrame) Dict[str, float]: [source]
Evaluates the model on the provided feature and target variables.
- predict_proba(x: pd.DataFrame) pd.DataFrame: [source]
Predicts the probabilities of outcomes for a set of instances.
- evaluate(X, y)[source]
Evaluates the model on the provided data.
@param X: The feature variables for evaluation, as a DataFrame. @param y: The target variable for evaluation, as a DataFrame. @return: A dictionary containing the loss and accuracy of the model.
- Return type:
Dict
[str
,float
]
- predict(X)[source]
Predicts outcomes for a set of instances.
@param X: The instances to predict, as a DataFrame. @return: Predictions as a DataFrame.
- Return type:
DataFrame
- predict_proba(x)[source]
Predicts the probabilities of outcomes for a set of instances.
@param x: The instances to predict, as a DataFrame. @return: Probabilities of each outcome as a DataFrame.
- Return type:
DataFrame
- predict_single(x)[source]
Predicts the outcome for a single instance.
@param x: The instance to predict, as a DataFrame. @return: The predicted class label (0 or 1).
- Return type:
int
- train(X, y, epochs=100, batch_size=32, **kwargs)[source]
Trains the model on the provided data.
@param X: The feature variables as a DataFrame. @param y: The target variable as a DataFrame. @param epochs: The number of epochs to train the model (default is 100). @param batch_size: The batch size used in training (default is 32).
- Return type:
None