asyml / asyml/forte

Investigate high-performance data serialization

Open
#405 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
253
Forks
59
PR merge metrics
No merged PRs in 30d

Description

This issue should be done at the end of project https://github.com/asyml/forte/projects/3.

**Is your feature request related to a problem? Please describe.**
Forte currently supports a simple serialization via JsonPickle. This format is fairly readable but the performance for loading and serialization are not optimized. Furthermore, we don't have control over the whole serialization process, which stops us from performing checking during the serialization

This issue needs to be further clarified after the main data structure is defined.

**Describe the solution you'd like**
1. Investigate the alternatives and create a rough estimation
2. Implement a different serialization that uses the new method.

**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.

**Additional context**
- This is part of the data efficiency project
- This PR should be made to the `new_datapack` branch.

Contributor guide

Open the contributing guide

Research direction

Start by reviewing the data efficiency project and the main data structure definition, then examine the current JsonPickle serialization path. Compare serialization alternatives and record a rough effort estimate before implementation. Done means a replacement serialization method is implemented on the new_datapack branch with control over the process and checking during serialization.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data-engineering
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.