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