I try to use torchtext library to help me handle a machine learning problem.
I use the simple data for an example, the goal is understand how to create the data object and dataloader before I start training model.
The csv file look like blow, there are only one column “text” and the element is sentence.
And I create a data object successfully.
After I build the vocab, I can use iterator to show mini-batch like blow.
I can understand the first column is index of words, but I cannot understand the second column, is it frequency of words?