Scribd Logo
Upload

Welcome to Scribd!

  • 43 views

    Lab 12

    Uploaded by

    Zameer Haider
    This document describes designing artificial neural networks to recognize gait activities using ankle-foot angle data. It provides training and target data for three gait activities: ramp ascend, ramp descend, and level ground. The methodology involves declaring variables, importing the excel data, creating a neural network using the nntool interface, training the network on the input-target data, and testing it on sample data. The output shows the network's predictions on test data samples and calculates the success rate.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Lab 12

    Uploaded by

    Zameer Haider
    43 views5 pages
    This document describes designing artificial neural networks to recognize gait activities using ankle-foot angle data. It provides training and target data for three gait activities: ramp ascend, ramp descend, and level ground. The methodology involves declaring variables, importing the excel data, creating a neural network using the nntool interface, training the network on the input-target data, and testing it on sample data. The output shows the network's predictions on test data samples and calculates the success rate.

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document describes designing artificial neural networks to recognize gait activities using ankle-foot angle data. It provides training and target data for three gait activities: ramp ascend, ramp descend, and level ground. The methodology involves declaring variables, importing the excel data, creating a neural network using the nntool interface, training the network on the input-target data, and testing it on sample data. The output shows the network's predictions on test data samples and calculates the success rate.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    43 views5 pages

    Lab 12

    Uploaded by

    Zameer Haider
    This document describes designing artificial neural networks to recognize gait activities using ankle-foot angle data. It provides training and target data for three gait activities: ramp ascend, ramp descend, and level ground. The methodology involves declaring variables, importing the excel data, creating a neural network using the nntool interface, training the network on the input-target data, and testing it on sample data. The output shows the network's predictions on test data samples and calculates the success rate.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    You are on page 1of 5

    Lab # 12

    Open Ended Lab


    Objective:
    Design artificial neural networks (ANN) to recognize gait activities using ankle-foot angles
    for various activities.
    Data:
    We are provided with training and target data (excel file) for three gait activities such as
    Gait Activities Target/Class
    Ramp Ascend (RA) 1
    Ramp Descend (RD) 2
    Level Ground (LG) 3

    We have two kinds of sample/testing data: 1) from training data, 2) external data.
    Artificial Neural Networks:
    An artificial neural network (ANN) is the piece of a computing system designed to
    simulate the way the human brain analyzes and processes information. It is the foundation
    of artificial intelligence (AI) and solves problems that would prove impossible or difficult by
    human or statistical standards.
    Uses:
    Artificial neural networks (ANN) are used for modelling non-linear problems and to
    predict the output values for given input parameters from their training values.
    Methodology:
    a) Variable Declaration:
    Declare variables and import data from the given excel file to the variables.

    Figure 1: variable declaration

    b) nntool Command:
    Write nntool command in the command window and enter. An interface will open and import all the
    training data, testing1, testing2 and Target data in it.
    Figure 2: nntool Interface

    Figure 3: Data import window

    c) Just click on the New tab button and enter the input output data, no. of neuron layers,
    functions and no. of neurons in the layers.

    Figure 4: Neural Network creation


    d) Then train the neural network for specific Input Targets data. By clicking on the Train
    Network button.

    Figure 6: Neural network training

    Figure 5: Regression values

    Figure 7: Output data

    e) Output:

    Sample Data 1 Network class


    (from training data) (1/2/3) prediction
    Col 1 2.8893
    Col 2 2.9302
    Col 3 2.0758
    Col 4 2.1578
    Col 5 1.0686
    Col 6 1.0423
    Success Rate 100%
    Testing data 2:

    Figure 8: Neural network creation

    Figure 9: Regression Values

    Figure 10: Output Values

    Output:

    Sample Data 1 Network class


    (from training (1/2/3) prediction
    data)
    Col 1 2.8233
    Col 2 1.4210
    Col 3 2.855
    Col 4 2.0257
    Col 5 1.5946
    Col 6 1.749
    Success Rate 66.6667%

    You might also like