Update merlin-core version specifier to match current release (#154)

update workflow on which docs pr preview depends (#141)

Co-authored-by: Oliver Holworthy <[email protected]>
Co-authored-by: Karl Higley <[email protected]>

adding ops-bot yaml file (#149)

Co-authored-by: Oliver Holworthy <[email protected]>

Use Dictionary lookup for supplied IDs to Embedding Operator (#148)

* Use lookup dict for Embedding operator with ids to speed up transform

* Add test for embedding operator with unknown value

* Update typehint for unknown_value

* Remove embedding tag from input schema to embeddings tests

The op now adds the embedding tag automatically, and the input schema
cannot have both a CATEGORICAL and EMBEDDING tag

* Flatten array passed  as ids to EmbeddingOperator

* Add assertion for ids shape

* Set default value of `embedding_index_mapping`

* Correct casing of message in test

Update base requirements for merlin 23.04

Warn users when a batch size may not work on GPU (#136)

* Warn users when a batch size may not work on GPU

* lint

Correct cpu-packages config - rename build -> build-packages (#109)

Replace `nnzs` with `row_lengths` for clarity (#99)

* Replace `nnzs` with `row_lengths` for clarity

The term `nnzs` (i.e. number of non-zeros) comes from sparse tensor nomenclature, but sparse tensors are an implementation detail of the dataloaders, not a domain concept we should be propagating throughout our code base. Instead, let's call them `row_lengths` which make sense as part of a ragged tensor representation throughout Merlin.

* Autoupdate precommit package versions

* Apply auto-formatting to appease the linter

Use tf.function for list column operations (#89)

add capability to target branch for unit test in cpu on tox (#58)

* add capability to target branch for unit test in cpu on tox

* fix cpu ci yml file name of next stage

* fix tox command in cpu ci