-
We'd like to integrate several varieties of negative sampling into the NVT/Merlin dataloaders:
- [ ] Popularity biased negatives from in-batch examples
- [ ] Uniform negatives from all items
- [ …
-
## Preamble
There ain't nobody that doesn't struggle with negatives, and it would be nice to have a linter that suggests a way to simplify boolean tests with negatives.
Here are a couple of ways…
-
Why don't you fine-tune the cross encoder for the tool retrieval task? Have you tried it? I have tried fine-tuning the cross encoder for the tool reranking task, but it performs very poorly, mostly re…
-
8/7, build 12978, vSphere 6.0 and 6.5
Nightly reports
```
13-1-vMotion-VCH-Appliance: | FAIL
```
However, all tests passed.
 or does it also work for B&W negatives?
…
-
The mediocre fuzzy matching etc. cause issues in the speaker matching algorithm. Especially false negatives seem to be an issue as we add more metadata.
- [x] Draw a sample of maybe 150 intros
- […
-
Analysis of simulated validation data by the CARD team (unpublished) revealed that Prodigal undercalls ORFs when the _-split_prodigal_jobs_ option is used. This was particularly noticeable in a _Acine…
-
It would be useful to allow negatives to handle cases where only some of the file's data is useful. For instance, if the timestamp on /foo is unimportant, but the checksum and permissions are importa…
-
# 🚀 Feature request
Is it possible to sample the negatives for the Two-Tower model from a column provided by the input data?
For example, we want to sample negatives from the list of items we disp…
-
**Describe the bug**
It is unclear to me if this is an issue when building or checking the Bloom filter; but either way, building a Bloom filter with `i8` or `i16` values (as opposed to `i32` or `i…