https://github.com/aaaastark/accelerometer-sensors-analysis

Time Series Analysis: Accelerometer Sensors of Object Inclination and Vibration. Time Series Analysis by using different (State of Art Models) Machine and Deep Learning. Recurent Neural Network with CuDNNLSTM Model, Convolutional Autoencoder, Residual Network (ResNet) and MobileNet Model.
https://github.com/aaaastark/accelerometer-sensors-analysis

accelerator accelerometer convolutional-autoencoder cudnnlstm inclination keras markov-transition-fields matplotlib moblienet pandas python recurrent-neural-network residual-network sensor sensors-data-collection skleran tensorflow time-series-analysis vibration-analysis vibration-sensor

Last synced: 3 months ago
JSON representation

Time Series Analysis: Accelerometer Sensors of Object Inclination and Vibration. Time Series Analysis by using different (State of Art Models) Machine and Deep Learning. Recurent Neural Network with CuDNNLSTM Model, Convolutional Autoencoder, Residual Network (ResNet) and MobileNet Model.

• Host: GitHub
• URL: https://github.com/aaaastark/accelerometer-sensors-analysis
• Owner: aaaastark
• License: mit
• Created: 2022-11-13T16:21:04.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2022-11-13T17:16:46.000Z (over 2 years ago)
• Last Synced: 2025-01-15T14:16:53.119Z (5 months ago)
• Topics: accelerator, accelerometer, convolutional-autoencoder, cudnnlstm, inclination, keras, markov-transition-fields, matplotlib, moblienet, pandas, python, recurrent-neural-network, residual-network, sensor, sensors-data-collection, skleran, tensorflow, time-series-analysis, vibration-analysis, vibration-sensor
• Homepage: https://github.com/aaaastark/Accelerometer-Sensors-Analysis
• Size: 16.6 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Time Series Analysis: Accelerometer Sensors of Object Inclination and Vibration

## Agenda: Time Series Analysis by using different (State of Art Models) Machine and Deep Learning.

 - Recurent Neural Network with CuDNNLSTM Model

 - Convolutional Autoencoder

 - Residual Network (ResNet) and MobileNet Model

#### Workflow that is used in this Project

- Data Processing/Transformation

- Data Normalization

- Image Transformation: Markov Transition Field

- State of Art Models (Machine and Deep Learning)

- Visulization of Train and Test Models: Accuracy and Loss 

- Classification Report, Accuracy, and Loss

#### APIs that are used in this Project

- tensorflow

- sklearn

- keras

- matplotlib

- numpy

- pandas

- pyts (Python Time Series Classification)

# Time Series Dataset: Accelerometer Sensors



    

    



# Image Transformation: Markov Transition Field



    

    



# Visulization of Train and Test Models: Accuracy and Loss



     

     

    

    



# Classification Report