I tried running a UnetClassifier model but i keep getting an error when trying to find the optimal learning rate. The error is as follows:
RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same
Now i found out that this means that the model is running on the gpu and my data is on the cpu. I used the prepare data function to create my data, and i can't find a option to get my data to prepare on GPU instead of CPU.
I am running my code in a notebook in arcgis pro version 3.0.2, the dataset i'm using is land_cover_classification_using_sparse_training_data and i'm running the code below
data = prepare_data(path=output_path, batch_size=6, imagery_type='naip', dataset_type='Classified_Tiles')
model = UnetClassifier(data, backbone='resnet34', ignore_classes=[0])
lr = model.lr_find()
