Instructions to use nvidia/dragon-multiturn-query-encoder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use nvidia/dragon-multiturn-query-encoder with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="nvidia/dragon-multiturn-query-encoder")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("nvidia/dragon-multiturn-query-encoder") model = AutoModel.from_pretrained("nvidia/dragon-multiturn-query-encoder") - Inference
- Notebooks
- Google Colab
- Kaggle
Reproduce evaluation result
#5
by r1ck - opened
I ran the evaluation using the script you provided, but I obtained a much higher score than what you reported in the paper. Here are the details:
doc2dial
top-1 recall score: 0.5011
top-5 recall score: 0.8385
top-20 recall score: 0.9533
quac
top-1 recall score: 0.6002
top-5 recall score: 0.8651
top-20 recall score: 0.9660