r9y9/deepvoice3_pytorch

Attention doesn't work well for downsample_step=1 and outputs_per_step=1

Open

#24 opened on Jan 10, 2018

View on GitHub
 (3 comments) (0 reactions) (0 assignees)Python (482 forks)github user discovery
bugenhancementhelp wanted

Repository metrics

Stars
 (1,980 stars)
PR merge metrics
 (PR metrics pending)

Description

Noticed while working on https://github.com/r9y9/deepvoice3_pytorch/pull/21.

Trained 300k steps, but the model was not generalized well. Need to figure out how we can improve.

Contributor guide

Research direction
Examine the attention mechanism implementation for downsample step=1 and outputs per step=1. Check gradient flow and alignment patterns in the training logs. Compare with working configurations (e.g., larger steps) to isolate the issue.
Tech stack
pythonpytorch
Domain
machine learningai
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Needs investigation
Prerequisites
PythonPyTorchdeep learning
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
40