facebookresearch/EmpatheticDialogues

Emotion Classifiers

Closed this issue · 9 comments

manzar96 commented

Hi,
I would like to ask in which dataset the classifier was pretrained and what was the architecture of the classifier.

EricMichaelSmith commented

Hi! A fastText classifier (https://fasttext.cc/) was trained out of the box to prepend emotion and topic labels.

manzar96 commented

Excuse me but I can't understand the concept exactly. The pretrained classifier was used to extract top k sentiments and the you prepended them to the input. (Am I right?) So that classifier was trained given the emotion labels of Empathetic Dataset or in another dataset?

EricMichaelSmith commented

I didn't train this particular classifier, but from the paper it looks like it was trained on the ED emotion labels: "To provide emotion signal, we train a classifier to predict the emotion label from the description of the situation written by the Speaker before the dialogue for the training set dialogues of ED (EMOPREPEND-1)."

manzar96 commented

ok! thank you a lot!

EricMichaelSmith commented

Sure thing :)

manzar96 commented

Can you please provide us the pre-trained classifier?

EricMichaelSmith commented

Hi! See notes for this in the README: do wget https://dl.fbaipublicfiles.com/parlai/empatheticdialogues/models/fasttext_empathetic_dialogues.mdl.

manzar96 commented

Thanks for your help! I would like to wrap this model and use it on my own code in another project (using pytorch). However I can not understand if that's possible as fasttext.load_model() returns a FastText object.
Can you please provide me some details to handle this issue?

EricMichaelSmith commented

Sorry for the delay! So, I've never tried porting FastText to PyTorch but I'm sure it's possible - Googling "fasttext to pytorch" gives lots of results that might help, for instance. Good luck!