Spaces:
Build error
Build error
File size: 1,376 Bytes
58cf517 30973d8 9de53c6 30973d8 9de53c6 4565d47 9de53c6 30973d8 9de53c6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 |
---
title: Augmented Poetry
emoji: ⚡
colorFrom: red
colorTo: gray
sdk: gradio
sdk_version: 3.4
app_file: app.py
pinned: false
---
- 1. fine-tune a large language model (LLM) using the text corpus of a specific poet
- select a certain rhyme from the gutenberg corpus and fine-tune on this
- try fine-tuning on a few lines of a poem that Eva has started
run in a docker container and transfer to another machine
Is it better to have a sequence to sequence transformer trained on sucessive lines of the poetry corpus??
https://github.com/huggingface/transformers/tree/main/examples/pytorch/language-modeling
merve/poetry only has 573 rows.
TODO: - upload the gutenberg poetry corpus up to huggingface - ask the lady who made it
## Research
<https://github.com/aparrish/gutenberg-dammit/>
implement language generation with a basic transformer
<https://github.com/aparrish/gutenberg-poetry-corpus>
Gutenberg Poetry Autocomplete, a search engine-like interface for writing poems mined from Project Gutenberg. (A poem written using this interface was recently published in the Indianapolis Review!)
https://ymeadows.com/en-articles/fine-tuning-transformer-based-language-models
https://thegradient.pub/prompting/
https://towardsdatascience.com/fine-tuning-for-domain-adaptation-in-nlp-c47def356fd6
https://ruder.io/recent-advances-lm-fine-tuning/
https://streamlit.io/ |