Open jeremytanjianle opened 4 years ago
Hi, absolutely love the repo and the paper. Congratulations on developing this.
Would just like to ask about the evaluation function in main.py
evaluation
Line 275: preds is derived from normalized_data_batch_1 but data_batch_2 is a different set of document altogether.
preds
data_batch_2
recon_loss = -(preds * data_batch_2).sum(1)
What does recon_loss mean if it matches the reconstruction of one set of documents against a different set of documents?
recon_loss
Same question here
IMHO data_batch_1 is a batch of first halve of documents and data_batch_2 is a batch of second halve of the same documents, namely document completion task.
data_batch_1
Hi, absolutely love the repo and the paper. Congratulations on developing this.
Would just like to ask about the
evaluation
function in main.pyLine 275:
preds
is derived from normalized_data_batch_1 butdata_batch_2
is a different set of document altogether.What does
recon_loss
mean if it matches the reconstruction of one set of documents against a different set of documents?