Model reference · open weights
gbert-germandpr-question_encoder is an open-weight embedding model from deepset. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Released by | deepset |
|---|---|
| Type | Embedding models |
| Task | Embeddings |
| Parameters (lead) | 110M |
| Context | 512 tokens |
| Runs with | transformers |
| Released | 2022-03-02 |
| Popularity | 705 downloads / month |
| Licence | Open weights |
About
Language model: gbert-base-germandpr Language: German Training data: GermanDPR train set (~ 56MB) Eval data: GermanDPR test set (~ 6MB) Infrastructure: 4x V100 GPU Published: Apr 26th, 2021
See https://deepset.ai/germanquad for more details and dataset download.
batch_size = 40
n_epochs = 20
num_training_steps = 4640
num_warmup_steps = 460
max_seq_len = 32 tokens for question encoder and 300 tokens for passage encoder
learning_rate = 1e-6
lr_schedule = LinearWarmup
embeds_dropout_prob = 0.1
num_hard_negatives = 2
During training, we monitored the in-batch average rank and the loss and evaluated different batch sizes, numbers of epochs, and number of hard negatives on a dev set split from the train set. The dev split contained 1030 question/answer pairs. Even without thorough hyperparameter tuning, we observed quite stable learning. Multiple restarts with different seeds produced quite similar results. Note that the in-batch average rank is influenced by settings for batch size and number of hard negatives. A smaller number of hard negatives makes the task easier. After fixing the hyperparameters we trained the model on the full GermanDPR train set.
We further evaluated the retrieval performance of the trained model on the full German Wikipedia with the GermanDPR test set as labels. To this end, we converted the GermanDPR test set to SQuAD format. The DPR model drastically outperforms the BM25 baseline with regard to recall@k.
You can load the model in haystack as a retriever for doing QA at scale:
retriever = DensePassageRetriever(
document_store=document_store,
query_embedding_model="deepset/gbert-base-germandpr-question_encoder"
passage_embedding_model="deepset/gbert-base-germandpr-ctx_encoder"
)
deepset is the company behind the production-ready open-source AI framework Haystack.
Some of our other work:
We also have a Discord community open to everyone!
Twitter | LinkedIn | Discord | GitHub Discussions | Website | YouTube
By the way: we're hiring!
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys gbert-germandpr-question-encoder for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (gbert-germandpr-question-encoder below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/embeddings \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"gbert-germandpr-question-encoder","input":"text to embed"}'
Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.