Skip to content

Conversation

Chandu-4444
Copy link
Contributor

Started working on text generation recipe for imdb dataset that uses the data in the unsup folder.

Chandu-4444 and others added 30 commits July 19, 2022 21:39
Minor changes as suggested by @ToucheSir
Change `AWD_LSTM` to `WeightDroppedLSTM`
@Chandu-4444
Copy link
Contributor Author

The next push contains all the changes from #250. I need that updated code from that branch to continue the work further.

Comment on lines 18 to 25
x_inp = map(i -> td[i][1][1:(end - 1)], 1:length(td))
y_out = map(i -> td[i][2][2:end], 1:length(td))

bv_x = BatchView(x_inp, batchsize = 4)
# bv_y = BatchView(y_out, batchsize = batch_size)
bv_y = BatchView(y_out, batchsize = 4)

return map((xs, ys) -> (batchseq(xs, 2), batchseq(ys, 2)), bv_x, bv_y)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

map -> mapobs, length -> numobs

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants