I’m reading the fastbook on my Ubuntu 18.04 laptop which is installed conda, Jupyter Notebook, CUDA, ipywidget, etc.
I’ve successfully executed the code cells in the chapter 02_production, starting from
# !pip install -Uqq fastbook
dls = bears.dataloaders(path)
(I think it should be noted that the code cell
im = Image.open(dest)
showed an image of a bear.)
showed no image, though it has executed something without error.
Same for the next four code cells that shows ResizeMethod.Squish, ResizeMethod.Pad, RandomResizedCrop, and aug_transforms.
I anyway executed until the cell
learn = cnn_learner(dls, resnet18, metrics=error_rate)
and after a while of running, the cell printed the table of epoch, train_loss, valid_loss, error_rate and time.
The next code cell,
interp = ClassificationInterpretation.from_learner(learn)
this doesn’t output anything. (Expected an image of confusion matrix)
Has anyone got into similar problem?
How did you fix it?