I am working on an Image Classification problem using Tensorflow and Keras. I am using a Tensorflow dataset (Batchdataset) The images belong to 9 classes. I need to check the number of images in each class. How do we check this? Do we convert the tf.dataset to a numpy array and then to a pandas dataframe and use value_counts()? Or is there an easier way to do it?
Asked
Active
Viewed 781 times
0
-
1[This](https://stackoverflow.com/questions/62878649/is-it-possible-to-get-class-distribution-when-using-tensorflow-datasets-tfds) one and [another one](https://stackoverflow.com/questions/65594438/how-to-plot-histogram-against-class-label-for-tf-dataset). I think they are what you are looking for. – Frightera Feb 21 '21 at 19:41