Tutorials

This section contains practical examples demonstrating the main features of nnodely through Jupyter notebooks.

All examples are provided without large datasets. Please refer to the repository for the complete data files.

Available Examples

• Load Data
  • Load a dataset using a directory
  • Load a dataset from a custom dictionary
  • Load a dataset from a pandas DataFrame
  • Resampling a pandas DataFrame
  • Get Samples from the Dataset
• Training
  • Basic Parameters and Usage
  • Recurrent Training
  • Set different weight for the minimization functions
  • Setting the Optimizer
  • Early stopping and Best Model
  • Train and Analyze
• Inference
  • Simple inference
  • Inference with a sampled data
  • Closed Loop and Connect during Inference
  • Recurrent inference with prediction samples
  • Recurrent inference with number of samples
• Import/Export Network
  • Save/Load the model definition in a json file
  • Save/Load network parameters in a pickle file
  • Export/Import pytorch model
  • Export the network to an ONNX format
  • ONNX inference
  • Export training and validation report
  • Example - Longitudinal dynamics of a vehicle (Non recurrent version export)
  • Example - Longitudinal dynamics of a vehicle (Recurrent version export)
• Fir Layer
  • Basic Usage Example
  • Setting a Weight and a Bias Name
  • Passing a Parameter to the Fir Relation
  • Initialize the parameters
  • Adding the Dropout
• Fuzzify Layer
  • Basic Usage
  • Rectangular function
  • Custom functions
  • Multiple custom functions
  • Repeated custom functions
• Linear Layer
  • Basic Usage
  • Passing a Parameter
  • Initialization functions
  • Dropout
• Local Model
  • Basic usage
  • Passing an output function
  • Passing a custom function
  • Custom function with multiple activations
  • Passing indexes
• Constant and Parameter
  • Basic Usage
  • Using a Parameter inside another block
  • Parametric functions
  • Arithmetic functions with Parameters and Constants
  • SampleTime
  • Initialization functions
• Parametric Functions
  • Basic usage
  • Parameter dimension
  • Passing custom parameters
  • Parametric function with different parameters
  • Parametric functions with Constants
  • Mapping over a batch
• Partitioning
  • Part
  • Select
  • Concatenate
  • SamplePart
  • SampleSelect
  • TimeConcatenate
• State Variables
  • State Definition
  • Closed Loop
  • Connect
  • Connect two models together
  • Recurrent Train
  • Clear State
  • State update at training time
• Equation Learner Layer
  • Basic Usage
  • Input layer
  • Input layer and output layer
  • Multiple inputs
  • Multi-parameter functions
  • Using custom parametric functions
  • Using parametric functions with parameters
  • Parametric functions and fuzzy layers
  • Cascade Equation Learner Blocks
• Interpolation Layer
  • Basic Usage
  • Linear Interpolation