NVIDIA-Merlin/dataloader issues and pull requests

#165 - Add GDPR to docs footers

Pull Request - State: open - Opened by mikemckiernan 7 months ago
Labels: documentation

#164 - fix for isaligned ragged failure #163

Pull Request - State: open - Opened by jperez999 about 1 year ago - 3 comments
Labels: bug

#163 - [BUG] Unable to extract session embeddings from a session-based transformer model

Issue - State: open - Opened by rnyak about 1 year ago - 3 comments
Labels: bug, P0

#162 - add rapids infra ci

Pull Request - State: closed - Opened by jperez999 about 1 year ago - 2 comments
Labels: ci

#161 - add rapids infra ci

Pull Request - State: closed - Opened by jperez999 about 1 year ago - 5 comments
Labels: ci

#160 - PyTorch Loader not working

Issue - State: closed - Opened by mohammad-qloo about 1 year ago - 2 comments

#159 - Use `copy-pr-bot`

Pull Request - State: closed - Opened by ajschmidt8 about 1 year ago - 2 comments

#158 - Out-of-memory error when iterating over merlin.dataloader.torch.Loader

Issue - State: closed - Opened by richtertill over 1 year ago - 1 comment

#157 - Update instructions for conda install

Pull Request - State: open - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: chore

#156 - Add fixture to cleanup dataloader after each test runs

Pull Request - State: open - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: chore

#155 - Specify merlin dependencies in setup.py to create release specifier

Pull Request - State: open - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: ci

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

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore, dependencies, skip-base-branch-check

#153 - Coerce Torch floating point values to default dtype

Pull Request - State: open - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: enhancement

#152 - Install dependencies in downstream tests

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: ci

#151 - Split gpu ci workflow into single-gpu and multi-gpu

Pull Request - State: open - Opened by edknv over 1 year ago - 3 comments
Labels: ci

#150 - Build docs for stable branch and make default

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: documentation

#149 - adding ops-bot yaml file

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: ci

#148 - Use Dictionary lookup for supplied IDs to Embedding Operator

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: enhancement

#147 - Make padding op faster and add error check

Pull Request - State: open - Opened by jperez999 over 1 year ago
Labels: enhancement

#146 - Adjust tag propagation in `EmbeddingOperator`

Pull Request - State: closed - Opened by karlhigley over 1 year ago
Labels: bug

#145 - Fix the output type of `Padding` op

Pull Request - State: closed - Opened by sararb over 1 year ago
Labels: enhancement

#144 - Remove use of deprecated numpy aliases of builtin types

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore

#143 - Update tag pattern in GitHub Workflows

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: ci

#142 - Add workflows to check base branch and set stable branch

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: ci

#141 - update workflow on which docs pr preview depends

Pull Request - State: closed - Opened by nv-alaiacano over 1 year ago - 1 comment
Labels: chore

#140 - Update requirements for Merlin packages to minimum version of 23.04

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: chore

#139 - fix array dl tests gpu vs cpu

Pull Request - State: closed - Opened by jperez999 over 1 year ago - 1 comment
Labels: bug, ci

#138 - Consolidate Embedding Operator and add offsets to embeddings

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: bug

#137 - Enable transforms to work when targets are output from dataloader

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore

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

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: chore

#135 - Apply device context when converting to torch tensors

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: bug

#134 - Revert "Set device for torch tensors with gpu > 1 (#132)"

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: bug, chore

#133 - removed base embedding operator, as it was a copy of the numpy op

Pull Request - State: closed - Opened by jperez999 over 1 year ago - 1 comment
Labels: chore

#132 - Set device for torch tensors with gpu > 1

Pull Request - State: closed - Opened by edknv over 1 year ago - 1 comment
Labels: bug

#131 - Device assignment does not work in PyTorch

Issue - State: open - Opened by edknv over 1 year ago - 4 comments

#130 - Dataloader does not work with tf.keras.layers.Embedding

Issue - State: closed - Opened by edknv over 1 year ago - 7 comments
Labels: bug, P0

#129 - Apply map function when peeking in tensorflow

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: bug

#128 - [FEA] Data loader: support to padding sparse sequential features on the left side

Issue - State: open - Opened by gabrielspmoreira about 3 years ago - 11 comments

#127 - fixing imports for tf and torch in dataloader

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: enhancement, chore

#126 - Add downstream repo tests

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: enhancement, ci

#125 - Adding Pad op

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: enhancement, chore

#124 - Feed pre-trained embeddings to NVTabular

Issue - State: open - Opened by MelissaKR over 2 years ago - 6 comments
Labels: question

#123 - Remove reset of dataloader in len method of `ArrayLoader`

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore

#122 - remove numpy tensorflow conversion logic in tf dataloader

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: bug, chore

#121 - Array loaderbase merge

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: chore

#120 - replace dictarray for tensortable in transforms functionality

Pull Request - State: closed - Opened by jperez999 over 1 year ago - 1 comment
Labels: chore

#119 - imports for GPU packages from compat

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: enhancement, ci, chore

#118 - Update checks of HAS_GPU in tests to handle case where cudf is not installed

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore

#117 - Optimize array dataloader performance

Pull Request - State: closed - Opened by karlhigley over 1 year ago

#116 - Remove reshape to speed up loading of scalar features in TensorFlow

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore

#115 - Change row lengths unit tests to use cpu and single gpu

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: bug, ci, chore

#114 - Update default value of cpu to None in dataset fixture

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore

#113 - Put row lengths on the same device on gpu

Pull Request - State: closed - Opened by edknv over 1 year ago
Labels: bug

#112 - Return Dense Tensors for Fixed List Columns

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: enhancement, breaking

#111 - adding array version of dataloader

Pull Request - State: closed - Opened by jperez999 over 1 year ago - 15 comments
Labels: enhancement

#110 - Apply the `EMBEDDING` tag to the output of the embedding operators

Pull Request - State: closed - Opened by karlhigley over 1 year ago
Labels: enhancement

#109 - Correct cpu-packages config

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: ci

#108 - GPU memory does not get freed up properly after each batch

Issue - State: open - Opened by felix0097 over 1 year ago - 5 comments
Labels: bug, P1

#107 - Restore support for passing `device` in CPU-only environment

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: chore

#106 - GPU is not detected properly when using SLURM

Issue - State: closed - Opened by felix0097 over 1 year ago - 5 comments

#105 - Change dataloader to output 1D tensors for scalar features

Issue - State: closed - Opened by viswa-nvidia over 1 year ago - 1 comment

#104 - Refactor ragged tensor processing for readbility and improved performance

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: enhancement

#103 - Remove sparse tensor output type for list features

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago
Labels: enhancement, breaking

#102 - docs: Rebuild docs on push

Pull Request - State: closed - Opened by mikemckiernan over 1 year ago
Labels: chore, skip-changelog

#101 - Update dataloader to provide new output structure

Pull Request - State: closed - Opened by bschifferer over 1 year ago - 6 comments
Labels: enhancement, breaking

#100 - make change to make transform occur on each batch as called

Pull Request - State: closed - Opened by jperez999 over 1 year ago - 4 comments
Labels: bug

#99 - Replace `nnzs` with `row_lengths` for clarity

Pull Request - State: closed - Opened by karlhigley over 1 year ago
Labels: chore

#98 - fix gpu visibility on docker

Pull Request - State: closed - Opened by jperez999 over 1 year ago
Labels: ci, chore

#97 - Improved Performance for Fixed List Columns

Pull Request - State: closed - Opened by oliverholworthy over 1 year ago - 1 comment
Labels: enhancement

#96 - Session based inference support, required changes for triton ragged support

Pull Request - State: closed - Opened by jperez999 almost 2 years ago - 1 comment
Labels: enhancement

#95 - Update Dataloader for `ColumnSchema` API changes from Merlin Core

Pull Request - State: closed - Opened by karlhigley almost 2 years ago
Labels: chore

#94 - Update docs support info in README

Pull Request - State: closed - Opened by mikemckiernan almost 2 years ago
Labels: chore, skip-changelog

#93 - Split CPU tests and building packages into separate Github Actions

Pull Request - State: closed - Opened by karlhigley almost 2 years ago
Labels: ci, chore

#92 - Add a legal statement to the notebooks

Pull Request - State: closed - Opened by mikemckiernan almost 2 years ago - 2 comments
Labels: documentation, skip-changelog

#91 - add "run in collab badge" to the PyTorch and TF examples

Pull Request - State: closed - Opened by radekosmulski almost 2 years ago - 3 comments
Labels: documentation, examples

#90 - add `map` to torch Loader

Pull Request - State: closed - Opened by radekosmulski almost 2 years ago - 1 comment
Labels: enhancement

#89 - Use tf.function for list column operations

Pull Request - State: closed - Opened by edknv almost 2 years ago - 1 comment
Labels: bug, ci

#88 - activate gha runner for gpu-ci

Pull Request - State: closed - Opened by jperez999 almost 2 years ago - 1 comment
Labels: ci, chore

#87 - Check Input Schema setter with list feature

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 3 comments
Labels: bug

#86 - Add schema setter to enable changing loader features on Loader instance

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 12 comments
Labels: enhancement

#85 - Shuffle doesn't work

Issue - State: open - Opened by Ilyushin almost 2 years ago - 3 comments
Labels: bug, P1

#84 - [Question] OOM Is there a way not to load the whole dataset in the dataloader?

Issue - State: closed - Opened by gaceladri almost 2 years ago - 1 comment

#83 - Remove echo from tox.ini

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 2 comments
Labels: ci, chore

#82 - NVTabular KerasSequenceLoader costs longer time to load multi-hot features than one-hot features

Issue - State: closed - Opened by Agoniii almost 2 years ago - 2 comments
Labels: bug, P0

#81 - add example for running on multiple GPUs using Horovod

Pull Request - State: closed - Opened by radekosmulski almost 2 years ago - 16 comments
Labels: documentation, examples

#80 - Support multiple label columns

Pull Request - State: closed - Opened by edknv almost 2 years ago - 3 comments
Labels: enhancement

#79 - Revert to using Keras Sequence for TensorFlow Loader

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 6 comments
Labels: bug

#78 - Adds context manager

Pull Request - State: closed - Opened by edknv almost 2 years ago - 2 comments
Labels: enhancement

#77 - Add `peek` method to get first batch without advancing iterator

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 3 comments
Labels: enhancement

#76 - [BUG] Dataloader doesnt release memory and memory growth

Issue - State: closed - Opened by bschifferer about 2 years ago - 6 comments
Labels: bug, question

#75 - [BUG] Data parallel training freezes due to different number of batches

Issue - State: open - Opened by bschifferer about 2 years ago - 17 comments
Labels: bug, P1

#74 - [BUG] Exception in model when using ragged tensors with tensorflow 2.10.0

Issue - State: closed - Opened by oliverholworthy about 2 years ago - 9 comments
Labels: bug, P1

#73 - Add Prettier formatter for YAML and Markdown

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 2 comments
Labels: chore

#72 - Does this work with images?

Issue - State: closed - Opened by fabiozappo almost 2 years ago - 1 comment

#71 - fix naming

Pull Request - State: closed - Opened by ArturKasymov almost 2 years ago - 3 comments
Labels: examples

#70 - Add common files to .gitignore

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 2 comments
Labels: chore

#68 - Add workflow to run tests with different TensorFlow versions

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 3 comments
Labels: ci

#66 - [Feature Request] Make the torch dataloader support TensorDict

Issue - State: open - Opened by vmoens almost 2 years ago
Labels: enhancement, question

#65 - Update TensorFlow Loader to run as iterator instead of Keras Sequence

Pull Request - State: closed - Opened by oliverholworthy almost 2 years ago - 2 comments
Labels: bug, breaking

#64 - [Task] Add multi-gpu data parallel example

Issue - State: closed - Opened by bschifferer almost 2 years ago
Labels: examples

GitHub / NVIDIA-Merlin/dataloader issues and pull requests