Comment on page
ReLU Layer

ReLU Layer Node Documentation
The ReLU Layer node in AugeLab Studio represents the Rectified Linear Unit (ReLU) activation layer for deep learning models.
Node Overview
The ReLU Layer node allows you to add a ReLU activation layer to a deep learning model. ReLU is a popular activation function that introduces non-linearity to the model.
Inputs
The ReLU Layer node does not require any inputs.
Outputs
The ReLU Layer node outputs the ReLU activation layer.
Node Interaction
1.Drag and drop the ReLU Layer node from the node library onto the canvas in AugeLab Studio.
2.Configure the node properties:
Maximum Value: Specify the maximum value for the ReLU activation. Any value above this threshold will be truncated.
Negative Side Slope: Specify the slope for the negative side of the ReLU activation function. A value of 0 will result in a standard ReLU activation.
Activation Threshold: Specify the threshold for the ReLU activation. Values below this threshold will be set to 0.
3.The ReLU activation layer will be created based on the specified configuration.
4.Connect the output ReLU activation layer to the appropriate nodes for building the deep learning model, such as the Add Layer node or the Model Building node.
Implementation Details
The ReLU Layer node is implemented as a subclass of the NodeCNN base class. It overrides the getKerasLayer method to create the ReLU activation layer.
The node reads the input values for the maximum value, negative side slope, and activation threshold.
The ReLU activation layer is created using the specified configuration.
The created layer is returned as the output.
Usage
1.Drag and drop the ReLU Layer node from the node library onto the canvas in AugeLab Studio.
2.Configure the node properties:
Maximum Value: Specify the maximum value for the ReLU activation. Any value above this threshold will be truncated.
Negative Side Slope: Specify the slope for the negative side of the ReLU activation function. A value of 0 will result in a standard ReLU activation.
Activation Threshold: Specify the threshold for the ReLU activation. Values below this threshold will be set to 0.
3.The ReLU activation layer will be created based on the specified configuration.
4.Connect the output ReLU activation layer to the appropriate nodes for building the deep learning model, such as the Add Layer node or the Model Building node.
5.Continue building the deep learning model by adding more layers or connecting other nodes as needed.
Notes
The ReLU Layer node allows you to add a ReLU activation layer to a deep learning model.
The ReLU activation function introduces non-linearity to the model, which is essential for capturing complex relationships in the data.
The maximum value parameter sets an upper limit for the ReLU activation. Any value above this threshold will be truncated to the maximum value.
The negative side slope parameter controls the slope of the activation function for negative input values. A value of 0 results in a standard ReLU activation.
The activation threshold parameter sets the threshold below which the ReLU activation function outputs 0. Values above the threshold are passed through without any modification.
Experiment with different parameter values to achieve the desired behavior for your deep learning model.
Connect the output ReLU activation layer to other nodes for building the deep learning model, such as the Add Layer node or the Model Building node.
Combine the ReLU activation layer with other layers and techniques to create powerful and expressive deep learning models.
The ReLU Layer node is suitable for various deep learning tasks, including image classification, object detection, and natural language processing.
Function Blocks - AI Model Elements - Previous
Optimizer SGD
Next - Function Blocks - AI Model Elements
Softmax Layer
Last modified 4mo ago