A Edge FPGA solution for classifying Tubercolosis and pnemonia in chest-xray images.
Model sized to be fit in the PYZNQ-2 Board
Folders:
Datasets ->Normal, Pneumonia and Tubercolosis (Around 2100 Images)
on_device -> Driver, test data and bitstream for on-device Deployment
cxr_cnn -> Vivado HLS project and Vivado project
NOTE:
-
HLS4ML Only supports Vivado 2019.02 as of now
-
A Bug in one of the tcl scripts (time variable overflow) requires you to set the system date to around 2019 before running hls_model.build()