/Incremental-Extreme-Learning-Machine-IELM

Implementation of Incremental Extreme Learning Machine [I-ELM] to deepen the understanding of the random network

Primary LanguagePython

Incremental-Extreme-Learning-Machine-IELM - Python

"Implementation of Incremental Extreme Learning Machine [I-ELM] in Python to deepen the understanding of the random network"

Part 1-A (Classification – ‘cifar10’ dataset)

Dataset:

‘cifar10’ - Dataset contains 60,000 32x32 color images in 10 different classes.

How to run:

-> Run the executable file ap_assignment-2_cifar10.py. -> Check the output.

Dataset and Program file structure:

‘ap_assignment-2_cifar10’: Parent file for reading dataset and applying I_ELM algorithm. ‘ielm_cifar10.py’: Defines I_ELM class along with classification training methods and testing methods.

Libraries:

1.) numpy: Python libraries multidimensional arrays and matrices.
2.) keras: Keras is an open-source neural-network library written in Python. Used for Dataset extraction.



Part 1-B (Classification – ‘mnist’ dataset)

Dataset:

‘mnist’ - Database of handwritten digits containing training set of 60,000 examples, and a test set of 10,000 examples.

How to run:

-> Run the executable file ap_assignment-2_mnist.py. -> Check the output.

Dataset and Program file structure:

‘ap_assignment-2_mnist’: Parent file for reading datasets and applying I_ELM algorithm.
‘ielm_mnist.py’: Defines I_ELM class along with classification training methods and testing methods.

Libraries:

1.) numpy: Python libraries multidimensional arrays and matrices.
2.) keras: Keras is an open-source neural-network library written in Python. Used for Dataset extraction.



Part 2 (Regression - Salary_Data)

Dataset:

Salary_Data - Database with salary details.

How to run:

-> Run the executable file ap_assignment-2_regression.py. -> Check the output.

Dataset and Program file structure:

‘ap_assignment-2_regression’: Parent file for reading dataset and applying of IELM for regression dataset.
‘ielm_regression.py’: Defines I_ELM method class, along with regression training methods and testing methods.
‘Salary_Data.csv’: Dataset downloaded from (https://archive.ics.uci.edu/ml/datasets)

Libraries:

1.) pandas: Reading data file which includes training as well as testing dataset.
2.) sklearn: For Splitting dataset.
3.) numpy: Python libraries for multidimensional arrays and matrices.