Commit
·
c9fa236
1
Parent(s):
aebc685
Update README.md
Browse files
README.md
CHANGED
|
@@ -15,3 +15,43 @@ Authors: *Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang
|
|
| 15 |
## Abstract
|
| 16 |
Transfer learning, where a model is first pre-trained on a data-rich task before being fine-tuned on a downstream task, has emerged as a powerful technique in natural language processing (NLP). The effectiveness of transfer learning has given rise to a diversity of approaches, methodology, and practice. In this paper, we explore the landscape of transfer learning techniques for NLP by introducing a unified framework that converts every language problem into a text-to-text format. Our systematic study compares pre-training objectives, architectures, unlabeled datasets, transfer approaches, and other factors on dozens of language understanding tasks. By combining the insights from our exploration with scale and our new “Colossal Clean Crawled Corpus”, we achieve state-of-the-art results on many benchmarks covering summarization, question answering, text classification, and more. To facilitate future work on transfer learning for NLP, we release our dataset, pre-trained models, and code.
|
| 17 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
## Abstract
|
| 16 |
Transfer learning, where a model is first pre-trained on a data-rich task before being fine-tuned on a downstream task, has emerged as a powerful technique in natural language processing (NLP). The effectiveness of transfer learning has given rise to a diversity of approaches, methodology, and practice. In this paper, we explore the landscape of transfer learning techniques for NLP by introducing a unified framework that converts every language problem into a text-to-text format. Our systematic study compares pre-training objectives, architectures, unlabeled datasets, transfer approaches, and other factors on dozens of language understanding tasks. By combining the insights from our exploration with scale and our new “Colossal Clean Crawled Corpus”, we achieve state-of-the-art results on many benchmarks covering summarization, question answering, text classification, and more. To facilitate future work on transfer learning for NLP, we release our dataset, pre-trained models, and code.
|
| 17 |
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
## Model series
|
| 22 |
+
This model is part of a series of models training on TPU with Flax Jax during Huggingface Flax/Jax challenge.
|
| 23 |
+
|
| 24 |
+
## Gpt models
|
| 25 |
+
|
| 26 |
+
## Swedish Gpt
|
| 27 |
+
https://huggingface.co/birgermoell/swedish-gpt/
|
| 28 |
+
|
| 29 |
+
## Swedish gpt wiki
|
| 30 |
+
https://huggingface.co/flax-community/swe-gpt-wiki
|
| 31 |
+
|
| 32 |
+
# Nordic gpt wiki
|
| 33 |
+
https://huggingface.co/flax-community/nordic-gpt-wiki
|
| 34 |
+
|
| 35 |
+
## Dansk gpt wiki
|
| 36 |
+
https://huggingface.co/flax-community/dansk-gpt-wiki
|
| 37 |
+
|
| 38 |
+
## Norsk gpt wiki
|
| 39 |
+
https://huggingface.co/flax-community/norsk-gpt-wiki
|
| 40 |
+
|
| 41 |
+
## Roberta models
|
| 42 |
+
|
| 43 |
+
## Nordic Roberta Wiki
|
| 44 |
+
https://huggingface.co/flax-community/nordic-roberta-wiki
|
| 45 |
+
|
| 46 |
+
## Swe Roberta Wiki Oscar
|
| 47 |
+
https://huggingface.co/flax-community/swe-roberta-wiki-oscar
|
| 48 |
+
|
| 49 |
+
## Roberta Swedish Scandi
|
| 50 |
+
https://huggingface.co/birgermoell/roberta-swedish-scandi
|
| 51 |
+
|
| 52 |
+
## Roberta Swedish
|
| 53 |
+
https://huggingface.co/birgermoell/roberta-swedish
|
| 54 |
+
|
| 55 |
+
## Swedish T5 model
|
| 56 |
+
https://huggingface.co/birgermoell/t5-base-swedish
|
| 57 |
+
|