Astur Apple image Dataset

Resumo

Struture of file: 2022-v2-9classes-AsturApple-Balanced-SIZE224-train-dev-test.hdf5<br> -------------------------------------------------------------------------------------------<br> This file contains dataset of 6108 cider apple color images, 224x224 pixels to support the article "Transfer learning with convolutional neural networks<br> for classification of cider apple varieties" results. -The images belong to nine apple classes: 'BLANQUINA' 'CARRIO' 'FLORINA' 'FUENTES' 'PRIETA' 'RAXAO' 'REINETA ENCARNADA' 'REINETA PINTA' 'REINETA ROJA DEL CANADA'<br> -Full dataset is split in 4886 images for training, 611 for testing and 611 for validation.<br> -Class labels are codified as one-hot enconding binary labels. (e.g. [0 0 0 0 0 1 0] ---&gt; Reineta Pinta)<br> -Training image set are store as tensor "trainX": (4168,224,224,3)<br> -Training class labels "trainY": (4886,7)<br> -Test image set tensor "testX": (611, 224,224,3)<br> -Test image binary class labels "testY": (611, 7)<br> -Validation image set tensor "devX": (611, 224,224,3)<br> -Validation binary class labels "devY": (611, 7) * file was created with h5py module in Python.