Jack Pertschuk

@Alex-Downey I would welcome a PR on this if you're able. Thanks

Fixed these I think.

could you post the body of the nboost response?

@jusjosgra Thank you for your positive feedback - I will try to improve the documentation. Essentially the way it works is that there are a set of config variables (which...

@jusjosgra all of these flags are **both** deploy time **and** dynamic, as you can **include it in the json body or query params** of the request to override. For example...

@kaykanloo @vchulski The numbers I posted are on a T4 GPU on Google Cloud. The numbers I see on the AWS p3.2xlarge should be most comparable to this I would...

are you running on multi-gpu environment?

if you run `curl localhost:8000/nboost/status` you should see some output like this: ```json {"avg_model_mrr":0.2475732343580853,"avg_num_choices":49.31506849315068,"avg_rerank_time":1.0792898919725527,"avg_response_time":0.6788859660548153,"avg_server_mrr":0.23740026696046881,"avg_topk":10.0} ``` The avg_num_choices is how many choices nboost gets back to rerank. The avg_topk is how...

@jishajoseph curl the nboost status endpoint as I described. The `avg_num_choices` from this represents the # returned by ES. So if this is only 5 , then there are no...

Do you mean train a reranking model? Or a model to extract answer spans from passages? In the former case I use the train [triples from say MS Marco](https://msmarco.blob.core.windows.net/msmarcoranking/triples.train.small.tar.gz) which...