dnrops
/
transformers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

correct RAG model cards (#7420)

This commit is contained in:
Patrick von Platen 2020-09-28 11:08:39 +02:00 committed by GitHub
parent 4083a55ab0
commit 8279471506
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 8 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
model_cards/facebook/rag-sequence-nq/README.md
View File

@ -11,17 +11,16 @@ by Patrick Lewis, Ethan Perez, Aleksandara Piktus et al.
The model is a *uncased* model, which means that capital letters are simply converted to lower-case letters.
The model consits of a *question_encoder*, *retriever* and a *generator*. The retriever is extracts relevant passages from the *wiki_dpr* `train` datasets, which is linked above.
The model consits of a *question_encoder*, *retriever* and a *generator*. The retriever extracts relevant passages from the *wiki_dpr* `train` datasets, which is linked above.
The question_encoder and retriever are based on `facebook/dpr-question_encoder-single-nq-base` and `facebook/bart-large`, which were jointly finetuned on
on the *wiki_dpr* QA dataset in an end-to-end fashion.
## Usage:
**Note**: In the usage example below only the *dummy* retriever of *wiki_dpr* is used because the real retriever requires to over 40 GB of RAM.
The model can generate questions to any question as follows:
**Note**: In the usage example below only the *dummy* retriever of *wiki_dpr* is used because the complete *lecagy* index requires over 75 GB of RAM.
The model can generate answers to any factoid question as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagSequenceForGeneration
tokenizer = RagTokenizer.from_pretrained("facebook/rag-token-nq")

2
model_cards/facebook/rag-token-base/README.md
View File

@ -29,6 +29,8 @@ Note that the model is *uncased* so that all capital input letters are converted
## Usage:
*Note*: the model uses the *dummy* retriever as a default. Better results are obtained by using the full retriever,
by setting `config.index_name="legacy"` and `config.use_dummy_dataset=False`.
The model can be fine-tuned as follows:
```python

6
model_cards/facebook/rag-token-nq/README.md
View File

@ -11,14 +11,14 @@ by Patrick Lewis, Ethan Perez, Aleksandara Piktus et al.
The model is a *uncased* model, which means that capital letters are simply converted to lower-case letters.
The model consits of a *question_encoder*, *retriever* and a *generator*. The retriever is extracts relevant passages from the *wiki_dpr* `train` datasets, which is linked above.
The model consits of a *question_encoder*, *retriever* and a *generator*. The retriever extracts relevant passages from the *wiki_dpr* `train` datasets, which is linked above.
The question_encoder and retriever are based on `facebook/dpr-question_encoder-single-nq-base` and `facebook/bart-large`, which were jointly finetuned on
on the *wiki_dpr* QA dataset in an end-to-end fashion.
## Usage:
**Note**: In the usage example below only the *dummy* retriever of *wiki_dpr* is used because the real retriever requires to over 40 GB of RAM.
The model can generate questions to any question as follows:
**Note**: In the usage example below only the *dummy* retriever of *wiki_dpr* is used because the complete *lecagy* index requires over 75 GB of RAM.
The model can generate answers to any factoid question as follows:
```python
from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration