Binary Image Classification with Tensorflow and EfficientNetB0 pre-trained model
And an other copy with Tensorflow and ResNet50 pre-trained model
This repo is a binary classification project that identify cells images as parasitized or uninfected cells from the thin blood smear slide images of segmented cells.
Python 3.x
Tensorflow 2.x
The Malaria dataset contains a total of 27,558 cell images with equal instances of parasitized and uninfected
size:337.08 MiB
On Tensorflow Datasets Catalog:Malaria
The National Library of medecin:NLM
EfficientNetB0 pre-trained model with 5,330,571 parameters
its size: 29 MB
Documentation
ResNet50 pre-trained model with 25,636,712 parameters
its size: 98 MB
Documentation
Cells segmentation with detectron2 and Mask R-CNN
Detectron2