TastyPancakes/Flux.jl
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

closes #411

Browse Source
This commit is contained in:
Mike J Innes 2018-10-03 11:45:29 +01:00 committed by GitHub
parent 3a7b77d104
commit fe6793fde5
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/layers/recurrent.jl
View File

@ -148,7 +148,7 @@ Base.show(io::IO, l::LSTMCell) =
print(io, "LSTMCell(", size(l.Wi, 2), ", ", size(l.Wi, 1)÷4, ")") print(io, "LSTMCell(", size(l.Wi, 2), ", ", size(l.Wi, 1)÷4, ")")
""" """
LSTM(in::Integer, out::Integer, σ = tanh) LSTM(in::Integer, out::Integer)
Long Short Term Memory recurrent layer. Behaves like an RNN but generally Long Short Term Memory recurrent layer. Behaves like an RNN but generally
exhibits a longer memory span over sequences. exhibits a longer memory span over sequences.
@ -189,7 +189,7 @@ Base.show(io::IO, l::GRUCell) =
print(io, "GRUCell(", size(l.Wi, 2), ", ", size(l.Wi, 1)÷3, ")") print(io, "GRUCell(", size(l.Wi, 2), ", ", size(l.Wi, 1)÷3, ")")
""" """
GRU(in::Integer, out::Integer, σ = tanh) GRU(in::Integer, out::Integer)
Gated Recurrent Unit layer. Behaves like an RNN but generally Gated Recurrent Unit layer. Behaves like an RNN but generally
exhibits a longer memory span over sequences. exhibits a longer memory span over sequences.