facebookresearch/fairseq

Inferring the decoder and encoder of a transformer model in a streaming manner

Open

#5,537 opened on Aug 29, 2024

View on GitHub
 (0 comments) (0 reactions) (0 assignees)Python (29,107 stars) (6,224 forks)batch import
enhancementhelp wantedneeds triage

Description

Is it possible to infer the model separately through encoder.onnx and decoder.onnx?

Contributor guide

Tech stack
pythonpytorch
Domain
machine learningai
Issue type
feature
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.
1-3 hours
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
familiarity with fairsequnderstanding of transformer modelsONNX runtime concepts
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
50
Research direction
Check fairseq's ONNX export capabilities by reviewing the documentation and examples in the repo. Investigate whether the encoder and decoder can be exported as separate ONNX graphs. Look for any existing issues or PRs related to streaming inference with ONNX in fairseq. Consider testing with a small model to verify separate inference.