From 95a2c40c8bbaf7014d4fc6e2da6480cc997306d0 Mon Sep 17 00:00:00 2001 From: Donny Greenberg Date: Wed, 1 Dec 2021 16:13:46 -0500 Subject: [PATCH 1/2] Create README.md --- README.md | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) create mode 100644 README.md diff --git a/README.md b/README.md new file mode 100644 index 000000000..7f946251e --- /dev/null +++ b/README.md @@ -0,0 +1,19 @@ +# TorchRec (🚨 Warning: Unstable Prototype 🚨) + +TorchRec is a PyTorch domain library built to provide common sparsity & parallelism primitives needed for large-scale recommender systems (RecSys). It allows authors to train models with large embedding tables sharded across many GPUs. + +## TorchRec contains: +- Parallelism primitives that enable easy authoring of large, performant multi-device/multi-node models using hybrid data-parallelism/model-parallelism. +- The TorchRec sharder can shard embedding tables with different sharding strategies including data-parallel, table-wise, row-wise, table-wise-row-wise, and column-wise sharding. +- The TorchRec planner can automatically generate optimized sharding plans for models. +- Pipelined training overlaps dataloading device transfer (copy to GPU), inter-device communications (input_dist), and computation (forward, backward) for increased performance. +- Optimized kernels for RecSys powered by FBGEMM. +- Quantization support for reduced precision training and inference. +- Common modules for RecSys. +- Production-proven model architectures for RecSys. +- RecSys datasets (criteo click logs and movielens) +- Examples of end-to-end training such the dlrm event prediction model trained on criteo click logs dataset. + +## Installation instructions +[Install FBGEMM] (https://github.com/pytorch/FBGEMM/tree/main/fbgemm_gpu#build-notes) +`pip install -e .` From 8a0736313b2bfab74d1241b202dbc99b08038dc5 Mon Sep 17 00:00:00 2001 From: Donny Greenberg Date: Wed, 1 Dec 2021 16:18:24 -0500 Subject: [PATCH 2/2] Delete README.MD --- torchrec/README.MD | 19 ------------------- 1 file changed, 19 deletions(-) delete mode 100644 torchrec/README.MD diff --git a/torchrec/README.MD b/torchrec/README.MD deleted file mode 100644 index 570db5b21..000000000 --- a/torchrec/README.MD +++ /dev/null @@ -1,19 +0,0 @@ -# Torchrec - -TorchRec is a PyTorch domain library built to provide common sparsity & parallelism primitives needed for large-scale recommender systems (RecSys). It allows authors to train models with large embedding tables sharded across many GPUs. - -## TorchRec contains: -- Parallelism primitives that enable easy authoring of large, performant multi-device/multi-node models using hybrid data-parallelism/model-parallelism. -- The TorchRec sharder can shard embedding tables with different sharding strategies including data-parallel, table-wise, row-wise, table-wise-row-wise, and column-wise sharding. -- The TorchRec planner can automatically generate optimized sharding plans for models. -- Pipelined training overlaps dataloading device transfer (copy to GPU), inter-device communications (input_dist), and computation (forward, backward) for increased performance. -- Optimized kernels for RecSys powered by FBGEMM. -- Quantization support for reduced precision training and inference. -- Common modules for RecSys. -- Production-proven model architectures for RecSys. -- RecSys datasets (criteo click logs and movielens) -- Examples of end-to-end training such the dlrm event prediction model trained on criteo click logs dataset. - -## Installation instructions -[Install FBGEMM] (https://github.com/pytorch/FBGEMM/tree/main/fbgemm_gpu#build-notes) -`pip install -e .`