Elias Bassani

Hi Paul, I agree with you, and I thought about it in the past. However, I plan to have a custom JSON-based file for `Run`s (_in addition to those already...

Added in `0.2.9`. Works with both `save` and `from_file`. `kind` now can be used to override the default behaviour.

Hi and thanks for the kind words! :D Kendall, Spearman, Pearson, and AP correlation metrics are already in my local branch and are coming soon! If you haven't, please consider...

Hi Stefano, Almost all the metrics ignore qrels with zero scores, including NDCG. So the difference you get is not because of that. However, I think your results are entirely...

Closing for inactivity. Feel free to re-open if needed.

@mohammed-elkomy and @jingtaozhan, thank you both for replying! I have further questions. What are `first_num` and `second_num`? Are they the _number of hard negative losses_ (one loss for each query-hard...

Thanks @mohammed-elkomy for the explanation. I probably swapped the random-negatives related code with the hard-negative one while reasoning about the implementation. Following your explanation, I assume `first_loss` is the [sum](https://github.com/jingtaozhan/DRhard/blob/1f7ba20798534459ba6106c1ee5c44eb110e0a5e/model.py#L153)...

Hi @Smu-Tan, @puzzlecollector, and @wise-east, I have just released a new Python-based search engine called [`retriv`](https://github.com/AmenRa/retriv). It only takes ~40ms to query 8M documents on my machine, and it can...

Hi, sorry for the delay. I will work on this as soon as I can.

Hi @ramsey-coding, I have just released a new Python-based search engine called [`retriv`](https://github.com/AmenRa/retriv). It only takes ~40ms to query 8M documents on my machine. If you try it, please, let...