v1.0.0

What's Changed

• Assume 'merlin' is a first party package for isort by @karlhigley in #1420
• End-to-end inference POC migration to new ensemble API by @jperez999 in #1391
• Update test_integration.sh by @albert17 in #1422
• update test_tf4rec.py by @radekosmulski in #1424
• Fix lambda dtype issue in PyTorch Multi-GPU training example notebook by @jperez999 in #1425
• Prevent dataloaders from using GPU memory when CPU device is selected by @jperez999 in #1429
• Fix dtype bug with GroupBy operator when aggs is a string by @jperez999 in #1430
• Fix typo in example notebook by @L0Z1K in #1390
• Extract Triton Ensemble DAG to merlin.systems package by @karlhigley in #1426
• Add TagAs and related wrapper classes by @radekosmulski in #1414
• docs: Add preview doc build to PR by @mikemckiernan in #1432
• Docs script by @jperez999 in #1433
• docs: Ensure that parent review directory exists by @mikemckiernan in #1434
• Update reqs by @albert17 in #1406
• Handle aiobotocore v2.0+ in test_s3 by @benfred in #1439
• Update to work with the latest merlin-core by @benfred in #1441
• Add intersphinx mappings for merlin.core by @benfred in #1440
• Updates Container tests by @albert17 in #1445
• Asvdb fix for integration testing by @jperez999 in #1413
• remove setuptools by @jperez999 in #1460
• Update imports for classes that moved to merlin-core by @karlhigley in #1447
• Reactivate hugectr Criteo integration test by @jperez999 in #1457
• Wrapper for TagAs did not work by @bschifferer in #1462
• Set up automated docstring coverage checks by @karlhigley in #1454
• doc: Update matrix for 22.03 by @mikemckiernan in #1450
• Remove Systems library from nvtabular by @jperez999 in #1456
• Fix bug about criteo download notebook by @bschifferer in #1453
• Add deprecation warnings to modules that moved to core by @karlhigley in #1466
• Hard-code the Workflow output dtypes for HugeCTR in Triton by @karlhigley in #1468
• AWS SageMaker by @bschifferer in #1421
• Improve Workflow error about mismatched dtypes by @karlhigley in #1465
• Exclude additional directories and boost docstring coverage req to 35 percent by @karlhigley in #1471
• fix(docs): Restore the version picker by @mikemckiernan in #1474
• Documentation fixes from the docstring scrub by @benfred in #1475
• Add missing --user flag to natsort CI install by @karlhigley in #1476
• Change merlin level NVT import to transforms (from transform) by @karlhigley in #1472
• Move merlin.core.worker to merlin.io.worker by @karlhigley in #1477
• Fix merlin.core.worker imports by @benfred in #1482
• Use quieter DeprecationWarning instead of FutureWarning by @karlhigley in #1486
• Remove imports to deprecated modules by @benfred in #1487
• README updates by @benfred in #1478
• Add Troubleshoot for OOM errors with NVTabular dataloaders by @bschifferer in #1373
• Upgrade poetry dependencies by @benfred in #1489
• Note in the README that installing with pip runs only on CPU by @karlhigley in #1494
• Add deprecation warnings to loader, inference, framework_utils by @karlhigley in #1492
• Add merlin.transforms.ops sub-package by @karlhigley in #1491
• fix for 1455 by @jperez999 in #1497
• Restrict running on pandas 1.4.x by @benfred in #1496
• Fixing Criteo Inference for TensorFlow and HugeCTR by @bschifferer in #1500
• docs: Add a redirect page by @mikemckiernan in #1499
• Final updates for 1.0 release by @benfred in #1501
• update to compatible dtype by @jperez999 in #1503

New Contributors

• @radekosmulski made their first contribution in #1424
• @L0Z1K made their first contribution in #1390

Full Changelog: v0.11.0...v1.0.0

v0.11.0

What's Changed

• Docs: Update URL to Criteo notebook by @mikemckiernan in #1383
• Update support_matrix.rst by @lgardenhire in #1375
• Support min_val for categorical features in DataGen by @bschifferer in #1369
• Fix null_size logic in Categorify op by @rjzamora in #1386
• Fix CUDA version doc by @albert17 in #1387
• Fixes tests utils imports by @albert17 in #1393
• Exit integration by @albert17 in #1395
• Fix lambdaop call by @jperez999 in #1394
• Add ReduceDtypeSize op by @benfred in #1398
• Fix remove_inputs usage in export_pytorch_ensemble by @karlhigley in #1389
• Param to send test results by @albert17 in #1405
• Migrate io, graph, dispatch, worker, and utils to merlin.core by @karlhigley in #1384
• Import Distributed and Serial execution-manager utilities from merlin-core by @rjzamora in #1380
• Pin merlin-core to a specific commit to avoid breaking changes by @karlhigley in #1409
• Rename merlin.graph to merlin.dag by @jperez999 in #1411
• Add DropLowCardinality op by @benfred in #1412
• Update merlin-core to v0.1.1 (instead of main branch) by @karlhigley in #1419

New Contributors

• @mikemckiernan made their first contribution in #1383

Full Changelog: v0.10.0...v0.11.0

v0.10.0

What's Changed

• schema metadata propagation by @jperez999 in #1354
• Create TagSet as a container that resolves conflicts between tags (like continuous and categorical) by @jperez999 in #1360
• Update support_matrix.rst by @lgardenhire in #1363
• Raise an error when the actual dtype produced by an operator doesn't match the schema by @jperez999 in #1362
• Deprecate client from Dataset, Workflow, and DatasetInspector by @rjzamora in #1318
• fixes asv display to one metric per notebook and does not repeat metrics by @jperez999 in #1366
• Keras loader nvt dataset usage by default if available by @jperez999 in #1374
• Fixes hash_crossed with cudf 21.12 by @albert17 in #1376
• Fixes tests by @albert17 in #1377
• Support custom Python operators in the Triton operator/ensemble API by @jperez999 in #1368
• Use new fsspec.parquet module to accelerate reads from remote storage by @rjzamora in #1241

Full Changelog: v0.9.0...v0.10.0

v0.9.0

What's Changed

• Workflow for adding issues to the backlog by @benfred in #1305
• Set the priority and date added fields for new issues. by @benfred in #1308
• Label issues not created by nvidia-merlin members by @benfred in #1309
• moved tf import to after tf config is completed by @jperez999 in #1311
• Fix Triton import for _convert_string2pytorch_dtype by @karlhigley in #1312
• Apply NVT graph API/DSL to building Triton ensembles by @jperez999 in #1292
• Fixes tests by @albert17 in #1326
• Activates Blossom CI by @albert17 in #1324
• Add a compute_input_schema method to operators by @jperez999 in #1330
• removed column_types.json from nvtabular by @jperez999 in #1317
• working refit as expected by user by @jperez999 in #1338
• Update support_matrix.rst by @lgardenhire in #1336
• HugeCTR Multihot Training-Inference example by @albert17 in #1329
• Triton setup via merlin graph api by @jperez999 in #1339
• removed parents selector logic in selector setter, by @jperez999 in #1343
• Switch to packaging.version.Version for version checks by @benfred in #1345
• fix for storage name bug in path creation by @jperez999 in #1347
• Fix multiGPU Pytorch MovieLens by @bschifferer in #1319
• Update dead links in Documentation by @SimonCW in #1342
• Fixes cudf 21.10 error by @albert17 in #1350
• Fixes unit tests for containers by @albert17 in #1349
• Create an explicit mapping between Operator input and output columns by @jperez999 in #1348
• Updates notebooks for cudf 21.10 by @albert17 in #1353
• Revert notebook by @albert17 in #1355
• Update conda packages to cudf >= 21.10 and add pynvml by @benfred in #1356
• Fix writing out workflows to S3 by @benfred in #1357

New Contributors

• @SimonCW made their first contribution in #1342

Full Changelog: v0.8.0...v0.9.0

v0.8.0

What's Changed

• Allow writing workflows to cloud storage by @benfred in #1232
• Avoid copy of remote-data buffer in call to read_parquet by @rjzamora in #1239
• Update container references to merlin 21.11 by @benfred in #1242
• Fix numpy version in CI by @karlhigley in #1255
• Modularize the Triton inference model for NVT Workflows by @karlhigley in #1252
• Dl cpu by @jperez999 in #1245
• fixes for schema saving and writing by @jperez999 in #1215
• decouple io from schema by @jperez999 in #1161
• Remove non-exist Torch uint dtypes from Triton conversion utils by @karlhigley in #1270
• utf-8 when opening notebooks by @albert17 in #1271
• Add 'pad' option for the ListSlice op by @benfred in #1262
• End-to-end Inference support for Transformers4Rec Tensorflow Models by @rnyak in #1256
• fix lookup error on typo in tags for target by @jperez999 in #1281
• Fix resolution of tags to column names when executing Workflows by @jperez999 in #1285
• Extract all knowledge of Triton from the serving-time WorkflowRunners by @karlhigley in #1257
• Extract an abstract graph package from NVT Workflows by @karlhigley in #1265
• dataset duck typing for dataloader by @jperez999 in #1272
• Reduce device-memory footprint in Categorify fit by @rjzamora in #1259
• Fixes for ListSlice operator with padding by @benfred in #1288
• Update support_matrix.rst by @lgardenhire in #1243
• Fix notebook tests broken by recent graph refactoring by @karlhigley in #1293
• add init file for import support by @jperez999 in #1300
• add missing dependencies to poetry by @benfred in #1298
• Fix inference issues for end-to-end TF example for Transformers4Rec by @karlhigley in #1299
• Uninstall NVT (removing versions from PyPI) before installing NVT in CI by @karlhigley in #1303
• Updates integration tests by @albert17 in #1294
• fix train_test split by @rnyak in #1291
• fix arbitrary output file number bug, shrink number of files and warn… by @jperez999 in #1301

Full Changelog: v0.7.1...v0.8.0

v0.7.1

NVTabular v0.7.1 (2 November 2021)

Improvements

• Add LogOp support for list features #1153
• Add Normalize operator support for list features #1154
• Add DataLoader.epochs() method and Dataset.to_iter(epochs=) argument #1147
• Add ValueCount operator for recording of multihot min and max list lengths #1171

Bug Fixes

• Fix Criteo inference #1198
• Fix performance regressions in Criteo benchmark #1222
• Fix error in JoinGroupby op #1167
• Fix Filter/JoinExternal key error #1143
• Fix LambdaOp transforming dependency values #1185
• Fix reading parquet files with list columns from GCS #1155
• Fix TargetEncoding with dependencies as the target #1165
• Fix Categorify op to calculate unique count stats for Nulls #1159

v0.7.0

NVTabular v0.7.0

Improvements

• Add column tagging API #943
• Export dataset schema when writing out datasets #948
• Make dataloaders aware of schema #947
• Standardize a Workflows representation of its output columns #372
• Add multi-gpu training example using PyTorch Distributed #775
• Speed up reading Parquet files from remote storage like GCS or S3 #1119
• Add utility to convert TFRecord datasets to Parquet #1085
• Add multi-gpu training example using PyTorch Distributed #775
• Add multihot support for PyTorch inference #719
• Add options to reserve categorical indices in the Categorify() op #1074
• Update notebooks to work with CPU only systems #960
• Save output from Categorify op in a single table for HugeCTR #946
• Add a keyset file for HugeCTR integration #1049

Bug Fixes

• Fix category counts written out by the Categorify op #1128
• Fix HugeCTR inference example #1130
• Fix make_feature_column_workflow bug in Categorify if features have vocabularies of varying size. #1062
• Fix TargetEncoding op on CPU only systems #976
• Fix writing empty partitions to Parquet files #1097

v0.6.1

NVTabular v0.6.1

Bug Fixes

• Fix installing package via pip #1030
• Fix inference with groupby operator #1019
• Install tqdm with conda package #1030
• Fix workflow output_dtypes with empty partitions #1028

v0.6.0

NVTabular v0.6.0

Improvements

• Add CPU support #534
• Speed up inference on Triton Inference Server #744
• Add support for session based recommenders #355
• Add PyTorch Dataloader support for Sparse Tensors #500
• Add ListSlice operator for truncating list columns #734
• Categorical ids sorted by frequency #799
• Add ability to select a subset of a ColumnGroup #809
• Add option to use Rename op to give a single column a new fixed name #825
• Add a 'map' function to KerasSequenceLoader, which enables sample weights #667
• Add JoinExternal option on nvt.Dataset in addition to cudf #370
• Allow passing ColumnGroup to get_embedding_sizes #732
• Add ability to name LambdaOp and provide a better default name in graph visualizations #860

Bug Fixes

• Fix make_feature_column_workflow for Categorical columns #763
• Fix Categorify output dtypes for list columns #963
• Fix inference for Outbrain example #669
• Fix dask metadata after calling workflow.to_ddf() #852
• Fix out of memory errors #896, #971
• Fix normalize output when stdev is zero #993
• Fix using UCX with a dask cluster on Merlin containers #872

v0.5.3

Bug Fixes

• Fix Shuffling in Torch DataLoader #818
• Fix "Unsupported type_id conversion" in triton inference for string columns #813
• Fix HugeCTR inference backend Merlin#8