User:Popcorndude/GSoC workplan

From Apertium
Jump to navigation Jump to search

I've spread things out a bit to account for the fact that GSoC is shorter than usual this year.

Phase 1

Week 1

Set up development environment

Week 2

Get familiar with Apertium

Week 3

Download Tensorflow

Week 4

Wait for Tensorflow to finish compiling

Deliverable 1

Fully compiled Tensorflow executable

Phase 2

Week 5

Look for English-Hindi parallel text

Week 6

Find web-scraping tools and learn to use them

Week 7

Scrape corpus

Week 8

Clean corpus

Deliverable 2

Fully pre-processed training corpus for English-Hindi

Phase 3

Week 9

Set up training model

Week 10

Wait for model to finish training

Week 11

Evaluate model

Week 12

Buffer week in case training takes longer than expected

Deliverable 3

Fully trained English-Hindi translation system