clab/dynet

Add cuDNN support for RNNs

Open

#574 opened on May 31, 2017

View on GitHub
 (0 comments) (0 reactions) (0 assignees)C++ (701 forks)github user discovery
enhancementhelp wanted

Repository metrics

Stars
 (3,433 stars)
PR merge metrics
 (PR metrics pending)

Description

It would be nice to have cuDNN speedup for SimpleRNN, LSTM and GRU.

While cuDNN will likely not be able to accelerate incremental RNN construction, we can special case the .transduce(seq) and .add_inputs(seq) implementations to make use of the acceleration.

Contributor guide

Research direction
Study the cuDNN RNN API and integrate it into DyNet's SimpleRNN, LSTM, and GRU implementations, focusing on the .transduce() and .add inputs() methods. Ensure compatibility with existing DyNet RNN abstractions.
Tech stack
cpp
Domain
machine learning
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
C++CUDAcuDNNDyNet internals
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
20