dsgibbons
dsgibbons
## Related Issues/PRs #6555 ## What changes are proposed in this pull request? 1. Store and use `original_column_names` when running inference in `default_evaluator` to avoid sklearn warning: `X has feature...
QAM currently only requires that the square root of the constellation size is an integer at initialization. QAM also requires that the constellation size be a power of 2. This...
Say I have a local filesystem with ``` . ├── a.yaml ├── b.yaml └── c └── d.yaml ``` Running `curl localhost:5000` currently outputs something like: ```bash HTTP File Server HTTP...
Closes #588 This PR takes a different course from what was proposed in #588 and the subsequent PR #605. Rather than using the existing `tag` field in the manifest, this...
When loading some models, I receive the WARN log: `Memory over-allocation due to under-prediction of model size...` (which stems from [here](https://github.com/kserve/modelmesh/blob/main/src/main/java/com/ibm/watson/modelmesh/ModelCacheUnloadBufManager.java#L233)) followed by the INFO log: `Eviction triggered for model...
Closes #85 Looks like some logic was copied between struct and list-of-structs validation. There were a few issues: - The optional outer was being removed before `is_optional`. - Inside the...
Closes #59 This PR adds a new field to `ColumnInfo` called `allow_missing`. During validation, if we find that a column is missing, we check if `allow_missing` is set inside `schema.column_infos[missing_column]`....
Closes #2200 I've tried to address all of the requirements from #2200, but I'd like some feedback to make sure the general approach is correct before I finish up. This...