Merlin
NVIDIA Merlin is an open source library providing end-to-end GPU-accelerated recommender systems, from feature engineering and preprocessing to training deep learning models and running inference in production.
파일 탐색기
최종 버전 다운로드 (.zip)- bug-report.md
- documentation-request.md
- feature-request.md
- release.md
- roadmap.md
- submit-question.md
- task.md
- check-base-branch.yaml
- docs-build.yaml
- docs-preview-pr.yaml
- docs-remove-stale-reviews.yaml
- docs-sched-rebuild.yaml
- gpu-ci.yml
- release-drafter.yaml
- require-label.yaml
- set-stable-branch.yaml
- triage.yml
- copy-pr-bot.yaml
- release-drafter.yml
- container_hugectr.sh
- container_integration.sh
- container_size.sh
- container_software.sh
- container_unit.sh
- dockerfile.ci
- pr.gpu.Jenkinsfile
- test_container.sh
- test_integration.sh
- dockerfile-nightly.merlin
- dockerfile.ctr
- dockerfile.merlin
- dockerfile.merlin.ctr
- dockerfile.tf
- dockerfile.torch
- hugectr-test.yaml
- merlin-test.yaml
- ngc_overview_hugectr.md
- ngc_overview_pytorch.md
- ngc_overview_tensorflow.md
- README.md
- tf-test.yaml
- torch-test.yaml
- data.json
- sample.json
- sample_after.json
- recommender-systems-dev-web-850.svg
- onetrust-loader.js
- favicon.png
- NVIDIA-LogoBlack.svg
- NVIDIA-LogoWhite.svg
- layout.html
- models_ranking.png
- models_retrieval.png
- models_sequentialinteactions.png
- models_sessionbased.png
- recommender_models.rst
- recommender_system_guide.rst
- index.rst
- support_matrix_merlin_hugectr.rst
- support_matrix_merlin_pytorch.rst
- support_matrix_merlin_tensorflow.rst
- conf.py
- containers.rst
- index.rst
- toc.yaml
- data.json
- extractor.py
- Makefile
- README.md
- requirements-doc.txt
- smx2rst.py
- snippets.yaml
- table_config.yaml
- test_extractor.py
- 01-Building-Recommender-Systems-with-Merlin.ipynb
- 02-Deploying-multi-stage-RecSys-with-Merlin-Systems.ipynb
- README.md
- gs-keras-model-plot.png
- triton-tf.png
- 01-Download-Convert.ipynb
- 02-ETL-with-NVTabular.ipynb
- 03-Training-with-HugeCTR.ipynb
- 03-Training-with-PyTorch.ipynb
- 03-Training-with-TF.ipynb
- 04-Triton-Inference-with-TF.ipynb
- README.md
- fourstages.png
- triton_ensemble.png
- transformers-next-item-prediction-with-pretrained-embeddings.ipynb
- transformers-next-item-prediction.ipynb
- mtl_mlp.png
- mtl_mmoe.png
- mtl_ple.png
- stl_dcn_click.png
- stl_deepfm_click.png
- stl_dlrm_click.png
- stl_mlp_click.png
- stl_wideanddeep_click.png
- hpo_evolution_deepfm.png
- hpo_parallel_bars_deepfm.png
- mtl_benchmark.png
- quick_start_process.png
- stl_benchmark.png
- tenrec_dataset.png
- wandb_sweeps.png
- __init__.py
- args_parsing.py
- inference.ipynb
- inference.py
- README.md
- __init__.py
- args_parsing.py
- preprocessing.py
- README.md
- mtl_mlp-stage01.yaml
- mtl_mlp-stage02.yaml
- mtl_mmoe-stage01.yaml
- mtl_mmoe-stage02.yaml
- mtl_ple-stage01.yaml
- mtl_ple-stage02.yaml
- stl_click_dcn.yaml
- stl_follow_dcn.yaml
- stl_like_dcn.yaml
- stl_share_dcn.yaml
- stl_click_deepfm.yaml
- stl_click_dlrm.yaml
- stl_follow_dlrm.yaml
- stl_like_dlrm.yaml
- stl_share_dlrm.yaml
- stl_click_mlp.yaml
- stl_follow_mlp.yaml
- stl_like_mlp.yaml
- stl_share_mlp.yaml
- stl_click_wide_n_deep.yaml
- README.md
- tutorial_with_wb_sweeps.md
- __init__.py
- args_parsing.py
- cufile.log
- mtl.py
- ranking.py
- ranking_models.py
- README.md
- __init__.py
- __init__.py
- ranking.md
- README.md
- requirements.txt
- Training-and-Deploying-DLRM-model-with-Models-and-Systems.ipynb
- README.md
- Dockerfile
- build_and_push_image.sh
- README.md
- sagemaker-merlin-tensorflow.ipynb
- train.py
- dask-dataframe.svg
- keras-model-plot.png
- triton-hugectr.png
- triton-tf.png
- 01-Download-Convert.ipynb
- 02-ETL-with-NVTabular.ipynb
- 03-Training-with-HugeCTR.ipynb
- 03-Training-with-Merlin-Models-TensorFlow.ipynb
- 04-Triton-Inference-with-Merlin-Models-TensorFlow.ipynb
- README.md
- Serving-An-Implicit-Model-With-Merlin-Systems.ipynb
- Serving-An-XGboost-Model-With-Merlin-Systems.ipynb
- README.md
- merlin_framework.png
- test_asvdb_transformers_next_item_prediction.py
- test_preproc.py
- test_ranking.py
- test_ci_building_deploying_multi_stage_RecSys.py
- test_serving_an_implicit_model_with_merlin_systems.py
- test_serving_an_xgboost_model_with_merlin_systems.py
- test_serving_ranking_models_with_merlin_systems.py
- test_version.cpython-38-pytest-7.1.0.pyc
- test_building_deploying_multi_stage_RecSys.py
- test_getting_started_hugectr.py
- test_getting_started_pytorch.py
- test_getting_started_tensorflow.py
- test_scaling_criteo_merlin_models.py
- test_scaling_criteo_merlin_models_hugectr.py
- test_scaling_criteo_optimize_notebook.py
- test_transformers_next_item_prediction.py
- test_transformers_next_item_prediction_with_pretrained_embeddings.py
- test_version.py
- __init__.py
- conftest.py
- .flake8
- .gitignore
- .pre-commit-config.yaml
- .pylintrc
- CHANGELOG.md
- CONTRIBUTING.md
- LICENSE
- pytest.ini
- README.md
- requirements.txt
- setup.cfg
- setup.py
- tox.ini
# 설치 가이드
1. 코드 내려받기
git clone https://github.com/NVIDIA-Merlin/Merlin
깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.
cd Merlin
방금 내려받은 프로젝트 폴더 안으로 이동합니다.
2. Docker
쉬움 추천사전 준비물
- Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
- Docker Desktop 컨테이너를 빌드하고 실행하려면 필요합니다. 설치 후 실행해서 백그라운드에 켜두세요.
docker build -f ci/dockerfile.ci -t merlin .
Dockerfile을 기반으로 실행 가능한 이미지를 빌드합니다.
docker run -p 8080:80 merlin
빌드된 이미지를 실제 컨테이너로 실행합니다.
터미널에 docker compose ps 를 입력해 컨테이너들이 Up 상태인지 확인하세요. README에 포트 번호가 적혀있다면 브라우저에서 http://localhost:포트번호 로 접속해보세요.
// repository documentation
Was this content helpful?
(0 ratings)
