Based on our paper "Pneumonia Detection from Lung X-ray Images using Local Search Aided Sine Cosine Algorithm based Deep Feature Selection Method" published in International Journal of Intelligent Systems, Wiley (2021).
Note: Access the preprint here. For the PDF of the published version of the manuscript, please email the first and/or second authors at: soumitri.chattopadhyay@gmail.com and rohitkunduju@gmail.com.
To install the required dependencies run the following in command prompt:
pip install -r requirements.txt
Required directory structure:
(Note: train and val contains subfolders representing classes in the dataset.)
+-- data
| +-- .
| +-- train
| +-- val
+-- AbSCA.py
+-- local_search.py
+-- main.py
Then, run the code using the command prompt as follows:
python main.py --data_directory "data"
Available arguments:
--epochs: Number of epochs of training. Default = 20--learning_rate: Learning Rate. Default = 0.001--batch_size: Batch Size. Default = 4--momentum: Momentum. Default = 0.9
If this article helps in your research in any way, please cite us using:
@article{chattopadhyay2022pneumonia,
title={Pneumonia detection from lung X-ray images using local search aided sine cosine algorithm based deep feature selection method},
author={Chattopadhyay, Soumitri and Kundu, Rohit and Singh, Pawan Kumar and Mirjalili, Seyedali and Sarkar, Ram},
journal={International Journal of Intelligent Systems},
DOI={10.1002/int.22703},
volume={37},
number={7},
pages={3777--3814},
year={2022},
publisher={Wiley Online Library}
}