For context, this is with regards to the Chexpert dataset.
I tried running the code before I realized that the way my data was stored was incorrect, preventing my training from running. I understand how to have my project directory based on the example given in the data folder, however I am uncertain how the images were originally pickled. For example, in the Chexpert dataset patient00005 has two studies, each of which have 2 images associated with it. How would you create the .npy file in that case?
To my current understanding I would use numpy.save() on the images in the dataset to create .npy files for each, however I want to make sure I have the correct understanding before I go and do so as the Chexpert dataset is massive and I have very limited data storage to recreate these results.
Any assistance or resources I could be pointed towards would be greatly appreciated.
For context, this is with regards to the Chexpert dataset.
I tried running the code before I realized that the way my data was stored was incorrect, preventing my training from running. I understand how to have my project directory based on the example given in the data folder, however I am uncertain how the images were originally pickled. For example, in the Chexpert dataset patient00005 has two studies, each of which have 2 images associated with it. How would you create the .npy file in that case?
To my current understanding I would use numpy.save() on the images in the dataset to create .npy files for each, however I want to make sure I have the correct understanding before I go and do so as the Chexpert dataset is massive and I have very limited data storage to recreate these results.
Any assistance or resources I could be pointed towards would be greatly appreciated.