MNIST image learner predict with fastai


I’ve been trying to build a neural network model for predicting the class (total 10 classes) of single-channel 28x28 images. I was successful in building and fitting the model but now I am encountering issues while trying to predict the class through using a 784 (28*28) length list or tensor.

How can I proceed to predict the class using the 784 length tensor. Is there any other method I should try for predictions? I’ve tried dataframe objects, series objects and numpy arrays but none of them seem to work.

Thank you