{"id":16392969,"url":"https://github.com/0h-n0/dl_benchmarks","last_synced_at":"2026-04-14T23:34:05.235Z","repository":{"id":129527964,"uuid":"109804101","full_name":"0h-n0/DL_benchmarks","owner":"0h-n0","description":"Simple benchmarks on major deeplearning frameworks with Python.","archived":false,"fork":false,"pushed_at":"2019-10-30T02:30:41.000Z","size":53,"stargazers_count":0,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-02-23T00:27:11.593Z","etag":null,"topics":["benchmark","chainer","cntk","deeplearning","mxnet","neon","pytorch","tensorflow"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/0h-n0.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2017-11-07T07:56:54.000Z","updated_at":"2019-10-30T02:30:43.000Z","dependencies_parsed_at":"2023-04-25T06:17:25.507Z","dependency_job_id":null,"html_url":"https://github.com/0h-n0/DL_benchmarks","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/0h-n0/DL_benchmarks","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0h-n0%2FDL_benchmarks","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0h-n0%2FDL_benchmarks/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0h-n0%2FDL_benchmarks/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0h-n0%2FDL_benchmarks/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/0h-n0","download_url":"https://codeload.github.com/0h-n0/DL_benchmarks/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0h-n0%2FDL_benchmarks/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31819891,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-14T18:05:02.291Z","status":"ssl_error","status_checked_at":"2026-04-14T18:05:01.765Z","response_time":153,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["benchmark","chainer","cntk","deeplearning","mxnet","neon","pytorch","tensorflow"],"created_at":"2024-10-11T04:51:53.886Z","updated_at":"2026-04-14T23:34:05.179Z","avatar_url":"https://github.com/0h-n0.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# DL_benchmarks WIP !!!\n\nDL_benchmarks surveys speed of each DeepLearning frameworks with dummy data.\nSo, Using dummy data, accuracy of results can not be compared. \n\n### Supported DeepLearning Frameworks.\n\n* tensorflow\n* tensorflow(Eagar)\n* tensorflow(Keras)\n* pytorch\n* chainer\n* mxnet\n* mxnet(gluon)\n* cntk\n* cntk(keras)\n* cntk(gluon?)\n* caffe(keras)\n* caffe2(python2...)\n* neon\n* tiny-net\n* nnabla\n* dynet\n* theano(keras)\n\n#### Requirements.\n\nSee requirements.txt.\n\n### Setup.\n\nI highly recommend using 'miniconda'. It is very easy to install a lot of DL frameworks with it.\nWithout 'miniconda', you must spend a lot of time to install them.\n\n```bash\n$ wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh\n$ # for linux\n$ sh Miniconda3-latest-Linux-x86_64.sh\n$ conda create -m benchmark pip\n$ source activate benchmark\n```\n\n```\n$ conda install pytorch torchvision cuda80 -c soumith\n$ # Your GPU is grater than 1080ti.\n$ # You should install pytorch from source.\n$ # conda install -c anaconda cmake\n$ # conda install -c conda-forge bzip2\n$ pip install mxnet-cu80\n$ # conda install -c pjmtdw mxnet-cudnn (cudnn-5~)\n$ pip install https://cntk.ai/PythonWheel/GPU/cntk-2.2-cp36-cp36m-linux_x86_64.whl\n$ pip install tensorflow-gpu\n$ git clone --recursive https://github.com/NervanaSystems/neon.git\n$ (cd neon \u0026\u0026 make sysinstall)\n$ pip install chainer cupy\n$ conda install -c conda-forge keras\n$ pip install cntk\n$ conda install -c mpi4py openmpi\n```\n\nSee setup.sh.\n\n\n#### build from source\n```\n$ # pytorch\n$ conda install -c anaconda cmake\n$ conda install -c conda-forge bzip2\n$ git clone --recursive https://github.com/pytorch/pytorch.git\n$ cd pytorch; python setup.py install\n$\n$ # chainer\n$ git clone --recursive https://github.com/chainer/chainer.git\n$ git clone --recursive https://github.com/chainer/chainer.git\n$ cd chainer; python setup.py install\n$ cd cupy; python setup.py install\n$\n$ # mxnet\n$ git clone --recursive https://github.com/apache/incubator-mxnet.git\n$ conda install -c intel mkl \n$ conda install -c intel/label/test mkl\n$ conda install -c intel/label/deprecated mkl\n$ conda install -c anaconda openblas \n$ \n$ $ git clone --recursive https://github.com/NervanaSystems/neon.git\n$ (cd neon \u0026\u0026 make sysinstall)\n$ pip install chainer cupy\n$ conda install -c conda-forge keras\n$ pip install cntk\n$ conda install -c mpi4py openmpi\n```\n\n#### How to use.\n\n```bash\n$ python -m benchmark.main\n$ \n$ python -m benchmark.main with framework=chainer\n$ # You can change framework.\n$\n$ python -m benchmark.main with framework=tensorflow data_config.batch_size=100\n$ \n$ python -m benchmark.main print_config\n$ # You can change configuration like the below code.\n$ # If you want to know more details about how to use it,\n$ # Plaese check sacred library and its homepage.\n```\n\n#### Result.\n\nWIP\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0h-n0%2Fdl_benchmarks","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F0h-n0%2Fdl_benchmarks","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0h-n0%2Fdl_benchmarks/lists"}