Open Beardlessape opened 5 years ago
Hello @Beardlessape. The reason for smaller dimensions is the feature selection (hence 5 audio features were useful). This can change from dataset to dataset and label to label. 20 is the maximum length of sequence we allow.
Thanks! but i still have some questions. @A2Zadeh How can i choose 20 of each sequence? Take the first twenty directly?
@Beardlessape the sequence is already in 20 format. (1284: data size,20: sequence length, 325: num features). Let us know if there is anything else.
Hi Paul, I have a question about the data ( X_train X_test X_valid)
for example, the X_train is (1284,20,325) and is the '325' composed of these three parts ,? 300d - text 20d - visual 5d - audio and what's the meaning of the '20'?