This example is to train a hematologic net model over the BloodMnist dataset.
The BloodMNIST , as a sub set of MedMNIST, is based on a dataset of individual normal cells, captured from individuals without infection, hematologic or oncologic disease and free of any pharmacologic treatment at the moment of blood collection. It contains a total of 17,092 images and is organized into 8 classes. it is split with a ratio of 7:1:2 into training, validation and test set. The source images with resolution 3×360×363 pixels are center-cropped into 3×200×200, and then resized into 3×28×28.
8 classes of the dataset:
"0": "basophil", "1": "eosinophil", "2": "erythroblast", "3": "ig (immature granulocytes)", "4": "lymphocyte", "5": "monocyte", "6": "neutrophil", "7": "platelet"
Download the pre-processed BloodMnist dataset to the folder (pathToDataset), which contains a few training samples and test samples. For the complete BloodMnist dataset, please download it via this link.
Start the training
python train.py hematologicnet -dir pathToDataset