I want to use my custom dataset to train a model, and it seems from here I'll have to have my dataset in the mentioned format.
Is there any way I can use my own custom dataloader to load the dataset in the pipeline?
Example: Suppose I have a dataset of numpy files which contains images in a numpy array.
I have followed the documentation here but was in vain.
Hi, first of all thanks a lot for this library.
I want to use my custom dataset to train a model, and it seems from here I'll have to have my dataset in the mentioned format.
Is there any way I can use my own custom dataloader to load the dataset in the pipeline? Example: Suppose I have a dataset of numpy files which contains images in a numpy array.
I have followed the documentation here but was in vain.
Any help would be really appreciated!
Thanks!