move merlin dependecy install in horovod-cpu tests (#1235)

* move merlin dependecy install in horovod-cpu tests

* trigger horovod test

Ensure TopKEncoder has correct outputs when model is saved (#1225)

* Remove output_names from base BaseModel.

* Add assertion for output signature of saved model to test_topk_encoder

* Move compile method from BaseModel to Model

* Correct name of structured outputs

* Move compile method back and add special case for TopKOutput

Makes RetrievalModelV2 support item tower with transforms (e.g. pre-t…

…rained embeddings) (#1198)

* Making retrieval model to_top_k_model(), candidate_embeddings() and batch_predict() support Loader with transforms for pre-trained embeddings in item tower

* Fixing test error and ensuring all batch_predict() with the new API support Loader with transforms (which include pre-trained embeddings)

* Fixing retrieval example, which was using wrong schema to export query and item embeddings

* Added missing importorskip on torch and pytorch_lightning for torch integration tests

* Skiping a test if nvtabular is available

Update merlin dependencies to match 23.06 release (#1155)

Add Model class (#1126)

* Add Model class

* Use BlockContainer in Models class

* Add unit tests

* Add docstrings

* Add module_utils unit tests

* Remove future work

* move initialize() to module_utils

* handle batch in training_step

* Add output_schema()

* check if model outputs have no targets when no target is provided

* put loss and metrics on the same device

* add docstrings to module_utils functions

* move metric device setting to initialize

* update logic for using model output targets

---------

Co-authored-by: Marc Romeyn <marcromeyn@gmail.com>

increase tolerance threshold (#1123)

Update requirements for merlin packages to minimum version of 23.04

MM transformer training and serving example (#1045)

Add inference to the next item prediction example (session-based)

* extend example
* align unit test
* add merlin-systems to envs

---------

Co-authored-by: Karl Higley <kmhigley@gmail.com>
Co-authored-by: Benedikt Schifferer <bschifferer@nvidia.com>

Increase tolerance in retrieval transformer test and random seed (#1007)

Update branch name extraction for tag builds (#972)

* Update branch name logic for tag builds to handle leading whitespace

* Add fetch for release branches