The dataset used for this project describes diagnosing of cardiac Single Proton Emission Computed Tomography (SPECT) images. Each of the patients is classified into two categories: normal(1) and abnormal(0).
The parallizing is done using OpenMP API.
SerialTraining.c contains the serial code which takes approximately 83 seconds to run 100000 epochs.
ParallelizedTraining.c contains the parallelized code that takes approximately 58 seconds to run 100000 epochs which shows the significant speed-up when comared to the corresponding serial code.