ContainersNVIDIADeepSeek-R1NVIDIALlama-3.1-Nemotron-70B-InstructNVIDIAPyTorch
CollectionsNVIDIAOmniverse Kit (FB)NVIDIADeepStream SDKNVIDIAOmniverse Kit App Streaming
ModelsNVIDIAStyleGAN3 pretrained modelsNVIDIAPeopleNetNVIDIATrafficCamNet
ResourcesNVIDIAKit SDK - Windows (PB25h1)NVIDIAKit SDK - Linux (PB25h1)NVIDIARiva Skills Quick Start
Helm ChartsNVIDIAGPU OperatorNVIDIANVIDIA NIM Operator
NVIDIA
NVIDIA
Merlin Training
Container
NVIDIA
NVIDIA
Merlin Training

This container allows users to do preprocessing and feature engineering with NVTabular, and then train a deep-learning based recommender system model with HugeCTR.

LayerLabelCreated
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4CONFIG
—
—
870475ceb3174479295984bd2061bae1df60d0356cc6fc00498a156ac820a8a4RUN
/bin/bash -c #(nop) ENTRYPOINT []
03/09/2021 6:15 AM UTC
f420f762ba2538e68c77d06a06a92f808b46311207e6c21d4d717c390d784916RUN
/bin/bash -c #(nop) HEALTHCHECK &{["NONE"] "0s" "0s" "0s" '\x00'}
03/09/2021 6:15 AM UTC
0691c403fc76c2462801f78ecdb4be91e554ee0702a6ffca8eedd4f073fe217eRUN
CC=8 CONDA_ENV=rapids CXX=8 HUGECTR_VER=v3.0 NVTAB_VER=v0.4.0 RAPIDS_VER=0.18 RAP_CHAN=rapidsai RELEASE=true /bin/bash -c echo $(du -h --max-depth=1 /)
03/09/2021 6:15 AM UTC
9f2553cf2b4b2d81be87d1fce11854259df666ef2503c7cd916ae43462ba7c0cRUN
CC=8 CONDA_ENV=rapids CXX=8 HUGECTR_VER=v3.0 NVTAB_VER=v0.4.0 RAPIDS_VER=0.18 RAP_CHAN=rapidsai RELEASE=true /bin/bash -c source activate ${CONDA_ENV}; conda clean --all -y
03/09/2021 6:15 AM UTC
8990c0cf6951c120cb4f13ce7bd70dd3e91186490774cf99d46a3e11e6122d7cRUN
CC=8 CONDA_ENV=rapids CXX=8 HUGECTR_VER=v3.0 NVTAB_VER=v0.4.0 RAPIDS_VER=0.18 RAP_CHAN=rapidsai RELEASE=true /bin/bash -c source activate ${CONDA_ENV}; conda env config vars set LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/hugectr/lib PATH=$PATH:/usr/local/hugectr/bin PYTHONPATH=$PYTHONPATH:/usr/local/hugectr/lib
03/09/2021 6:14 AM UTC
2f2ab051a5f6bd371d19d94cfc3c7b620dfd69a9a58e5559e9669efffcf7a554RUN
CC=8 CONDA_ENV=rapids CXX=8 HUGECTR_VER=v3.0 NVTAB_VER=v0.4.0 RAPIDS_VER=0.18 RAP_CHAN=rapidsai RELEASE=true /bin/bash -c source activate ${CONDA_ENV}; apt update; apt install -y graphviz ;
03/09/2021 6:14 AM UTC
53ce9dabc69f987184e27a3050342e9c9011a6f85df48d9f1deba468a9ff138eRUN
CC=8 CONDA_ENV=rapids CXX=8 HUGECTR_VER=v3.0 NVTAB_VER=v0.4.0 RAPIDS_VER=0.18 RAP_CHAN=rapidsai RELEASE=true /bin/bash -c rm /usr/local/cuda/lib64/stubs/libcuda.so.1
03/09/2021 6:14 AM UTC
d8ccd324a05936139a4722bc96371037d95a9965d9064a3462c490ffc2e33b5bRUN
CC=8 CONDA_ENV=rapids CXX=8 HUGECTR_VER=v3.0 NVTAB_VER=v0.4.0 RAPIDS_VER=0.18 RAP_CHAN=rapidsai RELEASE=true /bin/bash -c source activate ${CONDA_ENV}; cd /var/tmp/HugeCTR &&
  mkdir build &&
  cd build &&
  LD_LIBRARY_PATH=/usr/local/cuda/lib64/stubs/:$LD_LIBRARY_PATH &&
  export PATH=$PATH:/usr/local/cuda-11.0/compat/ &&
  cmake -DCMAKE_CXX_COMPILER=/usr/bin/g++ -DCMAKE_C_COMPILER=/usr/bin/gcc -DCMAKE_BUILD_TYPE=Release -DSM="60;61;70;75;80" -DVAL_MODE=OFF -DENABLE_MULTINODES=ON -DNCCL_A2A=ON .. &&
  make -j$(nproc) &&
  make install &&
  chmod +x /usr/local/hugectr/bin/* &&
  chmod +x /usr/local/hugectr/lib/* &&
  rm -rf /var/tmp/HugeCTR;
03/09/2021 6:14 AM UTC
5f8cbf096686410aae8043fb3523da61d85311ca7be67fe4ecf31500181c7a68RUN
CC=8 CONDA_ENV=rapids CXX=8 HUGECTR_VER=v3.0 NVTAB_VER=v0.4.0 RAPIDS_VER=0.18 RAP_CHAN=rapidsai RELEASE=true /bin/bash -c mkdir -p /var/tmp &&
  cd /var/tmp &&
  git clone https://github.com/NVIDIA/HugeCTR.git HugeCTR &&
  cd - &&
  cd /var/tmp/HugeCTR &&
  if [[ "$RELEASE" == "true" ]]; then git fetch --all --tags &&
  git checkout tags/${HUGECTR_VER}; else git checkout master; fi &&
  git submodule update --init --recursive
03/09/2021 6:04 AM UTC