Hi~
The output of the train_generator cell is the same as the expected output, but the output of the Test your function cell is as follows. Does anyone have any guesses as to why
You see the dimensions of targets and example_weights are wrong.
Since targets are labels for each input, which indicate if each input tweet is positive or negative, targets should be a list with n elements, where n is number of inputs.
And since example_weights are weights for each input, it should be also a list with n elements.
I think you can carefully debug your data_generator implementation.
I thought that the length of the len_data_pos and len_data_neg variables was the number of positive and negative labels, respectively, so the results were wrong. Thank you for your advice.