Is there a way to train a bidirectional RNN (like LSTM or GRU) on trax nowadays?
Contributor guide
Tech stack
python
Domain
machine learningdatabackend
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.
unclear
Prerequisites
Basic knowledge of TraxUnderstanding of RNNs
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
20
Research direction
Review the existing RNN implementation in Trax (e.g., trax/layers/rnn.py). Check the comments in the issue for discussion on how to add bidirectional support. Consider implementing a wrapper or modifying the RNN layer to accept a bidirectional flag. Look for any existing PRs or branches that attempt this.
Bidirectional RNN · google/trax#1078 | Good First Issue
